Abstract
Large language model (LLM)-based decision-making agents have shown the ability to generalize across multiple tasks. However, their performance relies on massive data and compute. We argue that this inefficiency stems from the forgetting phenomenon, in which a model memorizes its behaviors in parameters throughout training. As a result, training on a new task may deteriorate the model's performance on previous tasks. In contrast to LLMs' implicit memory mechanism, the human brain utilizes distributed memory storage, which helps manage and organize multiple skills efficiently, mitigating the forgetting phenomenon. Thus inspired, we propose an internal working memory module to store, blend, and retrieve information for different downstream tasks. Evaluation results show that the proposed method improves training efficiency and generalization in both Atari games and meta-world object manipulation tasks. Moreover, we demonstrate that memory fine-tuning further enhances the adaptability of the proposed architecture.
Differentiable Clustering with Perturbed Spanning Forests
Authors: Lawrence Stewart (SIERRA), Francis S Bach (SIERRA), Felipe Llinares López (IRIT), Quentin Berthet
Abstract
We introduce a differentiable clustering method based on minimum-weight spanning forests, a variant of spanning trees with several connected components. Our method relies on stochastic perturbations of solutions of linear programs, for smoothing and efficient gradient computations. This allows us to include clustering in end-to-end trainable pipelines. We show that our method performs well even in difficult settings, such as datasets with high noise and challenging geometries. We also formulate an ad hoc loss to efficiently learn from partial clustering data using this operation. We demonstrate its performance on several real world datasets for supervised and semi-supervised tasks.
Stecformer: Spatio-temporal Encoding Cascaded Transformer for Multivariate Long-term Time Series Forecasting
Abstract
Multivariate long-term time series forecasting is of great application across many domains, such as energy consumption and weather forecasting. With the development of transformer-based methods, the performance of multivariate long-term time series forecasting has been significantly improved, however, the study of spatial features extracting in transformer-based model is rare and the consistency of different prediction periods is unsatisfactory due to the large span. In this work, we propose a complete solution to address these problems in terms of feature extraction and target prediction. For extraction, we design an efficient spatio-temporal encoding extractor including a semi-adaptive graph to acquire sufficient spatio-temporal information. For prediction, we propose a Cascaded Decoding Predictor (CDP) to strengthen the correlation between different intervals, which can also be utilized as a generic component to improve the performance of transformer-based methods. The proposed method, termed as Spatio-temporal Encoding Cascaded Transformer (Stecformer), achieving a notable gap over the baseline model and is comparable with the state-of-the-art performance of transformer-based methods on five benchmark datasets. We hope our attempt will serve as a regular configuration in multivariate long-term time series forecasting in the future.
Abstract
Circuit representation learning aims to obtain neural representations of circuit elements and has emerged as a promising research direction that can be applied to various EDA and logic reasoning tasks. Existing solutions, such as DeepGate, have the potential to embed both circuit structural information and functional behavior. However, their capabilities are limited due to weak supervision or flawed model design, resulting in unsatisfactory performance in downstream tasks. In this paper, we introduce DeepGate2, a novel functionality-aware learning framework that significantly improves upon the original DeepGate solution in terms of both learning effectiveness and efficiency. Our approach involves using pairwise truth table differences between sampled logic gates as training supervision, along with a well-designed and scalable loss function that explicitly considers circuit functionality. Additionally, we consider inherent circuit characteristics and design an efficient one-round graph neural network (GNN), resulting in an order of magnitude faster learning speed than the original DeepGate solution. Experimental results demonstrate significant improvements in two practical downstream tasks: logic synthesis and Boolean satisfiability solving. The code is available at https://github.com/cure-lab/DeepGate2
Sim-Suction: Learning a Suction Grasp Policy for Cluttered Environments Using a Synthetic Benchmark
Abstract
This paper presents Sim-Suction, a robust object-aware suction grasp policy for mobile manipulation platforms with dynamic camera viewpoints, designed to pick up unknown objects from cluttered environments. Suction grasp policies typically employ data-driven approaches, necessitating large-scale, accurately-annotated suction grasp datasets. However, the generation of suction grasp datasets in cluttered environments remains underexplored, leaving uncertainties about the relationship between the object of interest and its surroundings. To address this, we propose a benchmark synthetic dataset, Sim-Suction-Dataset, comprising 500 cluttered environments with 3.2 million annotated suction grasp poses. The efficient Sim-Suction-Dataset generation process provides novel insights by combining analytical models with dynamic physical simulations to create fast and accurate suction grasp pose annotations. We introduce Sim-Suction-Pointnet to generate robust 6D suction grasp poses by learning point-wise affordances from the Sim-Suction-Dataset, leveraging the synergy of zero-shot text-to-segmentation. Real-world experiments for picking up all objects demonstrate that Sim-Suction-Pointnet achieves success rates of 96.76%, 94.23%, and 92.39% on cluttered level 1 objects (prismatic shape), cluttered level 2 objects (more complex geometry), and cluttered mixed objects, respectively. The Sim-Suction policies outperform state-of-the-art benchmarks tested by approximately 21% in cluttered mixed scenes.
Learning Better with Less: Effective Augmentation for Sample-Efficient Visual Reinforcement Learning
Authors: Guozheng Ma, Linrui Zhang, Haoyu Wang, Lu Li, Zilin Wang, Zhen Wang, Li Shen, Xueqian Wang, Dacheng Tao
Abstract
Data augmentation (DA) is a crucial technique for enhancing the sample efficiency of visual reinforcement learning (RL) algorithms. Notably, employing simple observation transformations alone can yield outstanding performance without extra auxiliary representation tasks or pre-trained encoders. However, it remains unclear which attributes of DA account for its effectiveness in achieving sample-efficient visual RL. To investigate this issue and further explore the potential of DA, this work conducts comprehensive experiments to assess the impact of DA's attributes on its efficacy and provides the following insights and improvements: (1) For individual DA operations, we reveal that both ample spatial diversity and slight hardness are indispensable. Building on this finding, we introduce Random PadResize (Rand PR), a new DA operation that offers abundant spatial diversity with minimal hardness. (2) For multi-type DA fusion schemes, the increased DA hardness and unstable data distribution result in the current fusion schemes being unable to achieve higher sample efficiency than their corresponding individual operations. Taking the non-stationary nature of RL into account, we propose a RL-tailored multi-type DA fusion scheme called Cycling Augmentation (CycAug), which performs periodic cycles of different DA operations to increase type diversity while maintaining data distribution consistency. Extensive evaluations on the DeepMind Control suite and CARLA driving simulator demonstrate that our methods achieve superior sample efficiency compared with the prior state-of-the-art methods.
SketchOGD: Memory-Efficient Continual Learning
Authors: Benjamin Wright, Youngjae Min, Jeremy Bernstein, Navid Azizan
Abstract
When machine learning models are trained continually on a sequence of tasks, they are liable to forget what they learned on previous tasks -- a phenomenon known as catastrophic forgetting. Proposed solutions to catastrophic forgetting tend to involve storing information about past tasks, meaning that memory usage is a chief consideration in determining their practicality. This paper proposes a memory-efficient solution to catastrophic forgetting, improving upon an established algorithm known as orthogonal gradient descent (OGD). OGD utilizes prior model gradients to find weight updates that preserve performance on prior datapoints. However, since the memory cost of storing prior model gradients grows with the runtime of the algorithm, OGD is ill-suited to continual learning over arbitrarily long time horizons. To address this problem, this paper proposes SketchOGD. SketchOGD employs an online sketching algorithm to compress model gradients as they are encountered into a matrix of a fixed, user-determined size. In contrast to existing memory-efficient variants of OGD, SketchOGD runs online without the need for advance knowledge of the total number of tasks, is simple to implement, and is more amenable to analysis. We provide theoretical guarantees on the approximation error of the relevant sketches under a novel metric suited to the downstream task of OGD. Experimentally, we find that SketchOGD tends to outperform current state-of-the-art variants of OGD given a fixed memory budget.
Learning Preconditioner for Conjugate Gradient PDE Solvers
Authors: Yichen Li, Peter Yichen Chen, Tao du, Wojciech Matusik
Abstract
Efficient numerical solvers for partial differential equations empower science and engineering. One of the commonly employed numerical solvers is the preconditioned conjugate gradient (PCG) algorithm which can solve large systems to a given precision level. One challenge in PCG solvers is the selection of preconditioners, as different problem-dependent systems can benefit from different preconditioners. We present a new method to introduce \emph{inductive bias} in preconditioning conjugate gradient algorithm. Given a system matrix and a set of solution vectors arise from an underlying distribution, we train a graph neural network to obtain an approximate decomposition to the system matrix to be used as a preconditioner in the context of PCG solvers. We conduct extensive experiments to demonstrate the efficacy and generalizability of our proposed approach in solving various 2D and 3D linear second-order PDEs.
Optimized Custom Dataset for Efficient Detection of Underwater Trash
Abstract
Accurately quantifying and removing submerged underwater waste plays a crucial role in safeguarding marine life and preserving the environment. While detecting floating and surface debris is relatively straightforward, quantifying submerged waste presents significant challenges due to factors like light refraction, absorption, suspended particles, and color distortion. This paper addresses these challenges by proposing the development of a custom dataset and an efficient detection approach for submerged marine debris. The dataset encompasses diverse underwater environments and incorporates annotations for precise labeling of debris instances. Ultimately, the primary objective of this custom dataset is to enhance the diversity of litter instances and improve their detection accuracy in deep submerged environments by leveraging state-of-the-art deep learning architectures.
Sample Efficient Reinforcement Learning in Mixed Systems through Augmented Samples and Its Applications to Queueing Networks
Authors: Honghao Wei, Xin Liu, Weina Wang, Lei Ying
Abstract
This paper considers a class of reinforcement learning problems, which involve systems with two types of states: stochastic and pseudo-stochastic. In such systems, stochastic states follow a stochastic transition kernel while the transitions of pseudo-stochastic states are deterministic given the stochastic states/transitions. We refer to such systems as mixed systems, which are widely used in various applications, including manufacturing systems, communication networks, and queueing networks. We propose a sample efficient RL method that accelerates learning by generating augmented data samples. The proposed algorithm is data-driven and learns the policy from data samples from both real and augmented samples. This method significantly improves learning by reducing the sample complexity such that the dataset only needs to have sufficient coverage of the stochastic states. We analyze the sample complexity of the proposed method under Fitted Q Iteration (FQI) and demonstrate that the optimality gap decreases as $\tilde{\mathcal{O}}(\sqrt{{1}/{n}}+\sqrt{{1}/{m}}),$ where $n$ is the number of real samples and $m$ is the number of augmented samples per real sample. It is important to note that without augmented samples, the optimality gap is $\tilde{\mathcal{O}}(1)$ due to insufficient data coverage of the pseudo-stochastic states. Our experimental results on multiple queueing network applications confirm that the proposed method indeed significantly accelerates learning in both deep Q-learning and deep policy gradient.
AD-NEV: A Scalable Multi-level Neuroevolution Framework for Multivariate Anomaly Detection
Authors: Marcin Pietron, Dominik Zurek, Kamil Faber, Roberto Corizzo
Abstract
Anomaly detection tools and methods present a key capability in modern cyberphysical and failure prediction systems. Despite the fast-paced development in deep learning architectures for anomaly detection, model optimization for a given dataset is a cumbersome and time consuming process. Neuroevolution could be an effective and efficient solution to this problem, as a fully automated search method for learning optimal neural networks, supporting both gradient and non-gradient fine tuning. However, existing methods mostly focus on optimizing model architectures without taking into account feature subspaces and model weights. In this work, we propose Anomaly Detection Neuroevolution (AD-NEv) - a scalable multi-level optimized neuroevolution framework for multivariate time series anomaly detection. The method represents a novel approach to synergically: i) optimize feature subspaces for an ensemble model based on the bagging technique; ii) optimize the model architecture of single anomaly detection models; iii) perform non-gradient fine-tuning of network weights. An extensive experimental evaluation on widely adopted multivariate anomaly detection benchmark datasets shows that the models extracted by AD-NEv outperform well-known deep learning architectures for anomaly detection. Moreover, results show that AD-NEv can perform the whole process efficiently, presenting high scalability when multiple GPUs are available.
Sliding Window Sum Algorithms for Deep Neural Networks
Authors: Roman Snytsar
Subjects: Machine Learning (cs.LG); Data Structures and Algorithms (cs.DS)
Abstract
Sliding window sums are widely used for string indexing, hashing and time series analysis. We have developed a family of the generic vectorized sliding sum algorithms that provide speedup of O(P/w) for window size $w$ and number of processors P. For a sum with a commutative operator the speedup is improved to O(P/log(w)). Even more important, our algorithms exhibit efficient memory access patterns. In this paper we study the application of the sliding sum algorithms to the training and inference of the Deep Neural Networks. We demonstrate how both pooling and convolution primitives could be expressed as sliding sums and evaluated by the compute kernels with the shared structure. We show that the sliding sum convolution kernels are more efficient than the commonly used GEMM kernels on the CPU, and could even outperform their GPU counterparts.
Abstract
We present efficient algorithms for Quantile Join Queries, abbreviated as %JQ. A %JQ asks for the answer at a specified relative position (e.g., 50% for the median) under some ordering over the answers to a Join Query (JQ). Our goal is to avoid materializing the set of all join answers, and to achieve quasilinear time in the size of the database, regardless of the total number of answers. A recent dichotomy result rules out the existence of such an algorithm for a general family of queries and orders. Specifically, for acyclic JQs without self-joins, the problem becomes intractable for ordering by sum whenever we join more than two relations (and these joins are not trivial intersections). Moreover, even for basic ranking functions beyond sum, such as min or max over different attributes, so far it is not known whether there is any nontrivial tractable %JQ. In this work, we develop a new approach to solving %JQ. Our solution uses two subroutines: The first one needs to select what we call a "pivot answer". The second subroutine partitions the space of query answers according to this pivot, and continues searching in one partition that is represented as new %JQ over a new database. For pivot selection, we develop an algorithm that works for a large class of ranking functions that are appropriately monotone. The second subroutine requires a customized construction for the specific ranking function at hand. We show the benefit and generality of our approach by using it to establish several new complexity results. First, we prove the tractability of min and max for all acyclic JQs, thereby resolving the above question. Second, we extend the previous %JQ dichotomy for sum to all partial sums. Third, we handle the intractable cases of sum by devising a deterministic approximation scheme that applies to every acyclic JQ.
CARAMEL: A Succinct Read-Only Lookup Table via Compressed Static Functions
Authors: Benjamin Coleman, David Torres Ramos, Vihan Lakshman, Chen Luo, Anshumali Shrivastava
Subjects: Data Structures and Algorithms (cs.DS); Databases (cs.DB); Information Retrieval (cs.IR)
Abstract
Lookup tables are a fundamental structure in many data processing and systems applications. Examples include tokenized text in NLP, quantized embedding collections in recommendation systems, integer sketches for streaming data, and hash-based string representations in genomics. With the increasing size of web-scale data, such applications often require compression techniques that support fast random $O(1)$ lookup of individual parameters directly on the compressed data (i.e. without blockwise decompression in RAM). While the community has proposd a number of succinct data structures that support queries over compressed representations, these approaches do not fully leverage the low-entropy structure prevalent in real-world workloads to reduce space. Inspired by recent advances in static function construction techniques, we propose a space-efficient representation of immutable key-value data, called CARAMEL, specifically designed for the case where the values are multi-sets. By carefully combining multiple compressed static functions, CARAMEL occupies space proportional to the data entropy with low memory overheads and minimal lookup costs. We demonstrate 1.25-16x compression on practical lookup tasks drawn from real-world systems, improving upon established techniques, including a production-grade read-only database widely used for development within Amazon.com.
CVB: A Video Dataset of Cattle Visual Behaviors
Authors: Ali Zia, Renuka Sharma, Reza Arablouei, Greg Bishop-Hurley, Jody McNally, Neil Bagnall, Vivien Rolland, Brano Kusy, Lars Petersson, Aaron Ingham
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Existing image/video datasets for cattle behavior recognition are mostly small, lack well-defined labels, or are collected in unrealistic controlled environments. This limits the utility of machine learning (ML) models learned from them. Therefore, we introduce a new dataset, called Cattle Visual Behaviors (CVB), that consists of 502 video clips, each fifteen seconds long, captured in natural lighting conditions, and annotated with eleven visually perceptible behaviors of grazing cattle. We use the Computer Vision Annotation Tool (CVAT) to collect our annotations. To make the procedure more efficient, we perform an initial detection and tracking of cattle in the videos using appropriate pre-trained models. The results are corrected by domain experts along with cattle behavior labeling in CVAT. The pre-hoc detection and tracking step significantly reduces the manual annotation time and effort. Moreover, we convert CVB to the atomic visual action (AVA) format and train and evaluate the popular SlowFast action recognition model on it. The associated preliminary results confirm that we can localize the cattle and recognize their frequently occurring behaviors with confidence. By creating and sharing CVB, our aim is to develop improved models capable of recognizing all important behaviors accurately and to assist other researchers and practitioners in developing and evaluating new ML models for cattle behavior classification using video data.
ParaAMR: A Large-Scale Syntactically Diverse Paraphrase Dataset by AMR Back-Translation
Abstract
Paraphrase generation is a long-standing task in natural language processing (NLP). Supervised paraphrase generation models, which rely on human-annotated paraphrase pairs, are cost-inefficient and hard to scale up. On the other hand, automatically annotated paraphrase pairs (e.g., by machine back-translation), usually suffer from the lack of syntactic diversity -- the generated paraphrase sentences are very similar to the source sentences in terms of syntax. In this work, we present ParaAMR, a large-scale syntactically diverse paraphrase dataset created by abstract meaning representation back-translation. Our quantitative analysis, qualitative examples, and human evaluation demonstrate that the paraphrases of ParaAMR are syntactically more diverse compared to existing large-scale paraphrase datasets while preserving good semantic similarity. In addition, we show that ParaAMR can be used to improve on three NLP tasks: learning sentence embeddings, syntactically controlled paraphrase generation, and data augmentation for few-shot learning. Our results thus showcase the potential of ParaAMR for improving various NLP applications.
Legion: Automatically Pushing the Envelope of Multi-GPU System for Billion-Scale GNN Training
Authors: Jie Sun, Li Su, Zuocheng Shi, Wenting Shen, Zeke Wang, Lei Wang, Jie Zhang, Yong Li, Wenyuan Yu, Jingren Zhou, Fei Wu
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract
Graph neural network(GNN) has been widely applied in real-world applications, such as product recommendation in e-commerce platforms and risk control in financial management systems. Several cache-based GNN systems have been built to accelerate GNN training in a single machine with multiple GPUs. However, these systems fail to train billion-scale graphs efficiently, which is a common challenge in the industry. In this work, we propose Legion, a system that automatically pushes the envelope of multi-GPU systems for accelerating billion-scale GNN training. First, we design a hierarchical graph partitioning mechanism that significantly improves the multi-GPU cache performance. Second, we build a unified multi-GPU cache that helps to minimize the PCIe traffic incurred by caching both graph topology and features with the highest hotness. Third, we develop an automatic caching management mechanism that adapts the multi-GPU cache plan according to the hardware specifications and various graphs to maximize the overall training throughput. Evaluations on various GNN models and multiple datasets show that Legion supports training billion-scale GNNs in a single machine and significantly outperforms the state-of-the-art cache-based systems on small graphs.
Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models
Abstract
Parameter-efficient tuning (PET) methods fit pre-trained language models (PLMs) to downstream tasks by either computing a small compressed update for a subset of model parameters, or appending and fine-tuning a small number of new model parameters to the pre-trained network. Hand-designed PET architectures from the literature perform well in practice, but have the potential to be improved via automated neural architecture search (NAS). We propose an efficient NAS method for learning PET architectures via structured and unstructured pruning. We present experiments on GLUE demonstrating the effectiveness of our algorithm and discuss how PET architectural design choices affect performance in practice.
Set-based Neural Network Encoding
Authors: Bruno Andreis, Soro Bedionita, Sung Ju Hwang
Abstract
We propose an approach to neural network weight encoding for generalization performance prediction that utilizes set-to-set and set-to-vector functions to efficiently encode neural network parameters. Our approach is capable of encoding neural networks in a modelzoo of mixed architecture and different parameter sizes as opposed to previous approaches that require custom encoding models for different architectures. Furthermore, our \textbf{S}et-based \textbf{N}eural network \textbf{E}ncoder (SNE) takes into consideration the hierarchical computational structure of neural networks by utilizing a layer-wise encoding scheme that culminates to encoding all layer-wise encodings to obtain the neural network encoding vector. Additionally, we introduce a \textit{pad-chunk-encode} pipeline to efficiently encode neural network layers that is adjustable to computational and memory constraints. We also introduce two new tasks for neural network generalization performance prediction: cross-dataset and cross-architecture. In cross-dataset performance prediction, we evaluate how well performance predictors generalize across modelzoos trained on different datasets but of the same architecture. In cross-architecture performance prediction, we evaluate how well generalization performance predictors transfer to modelzoos of different architecture. Experimentally, we show that SNE outperforms the relevant baselines on the cross-dataset task and provide the first set of results on the cross-architecture task.
FARA: Future-aware Ranking Algorithm for Fairness Optimization
Authors: Tao Yang, Zhichao Xu, Zhenduo Wang, Qingyao Ai
Abstract
Ranking systems are the key components of modern Information Retrieval (IR) applications, such as search engines and recommender systems. Besides the ranking relevance to users, the exposure fairness to item providers has also been considered an important factor in ranking optimization. Many fair ranking algorithms have been proposed to jointly optimize both ranking relevance and fairness. However, we find that most existing fair ranking methods adopt greedy algorithms that only optimize rankings for the next immediate session or request. As shown in this paper, such a myopic paradigm could limit the upper bound of ranking optimization and lead to suboptimal performance in the long term. To this end, we propose FARA, a novel Future-Aware Ranking Algorithm for ranking relevance and fairness optimization. Instead of greedily optimizing rankings for the next immediate session, FARA plans ahead by jointly optimizing multiple ranklists together and saving them for future sessions. Particularly, FARA first uses the Taylor expansion to investigate how future ranklists will influence the overall fairness of the system. Then, based on the analysis of the Taylor expansion, FARA adopts a two-phase optimization algorithm where we first solve an optimal future exposure planning problem and then construct the optimal ranklists according to the optimal future exposure planning. Theoretically, we show that FARA is optimal for ranking relevance and fairness joint optimization. Empirically, our extensive experiments on three semi-synthesized datasets show that FARA is efficient, effective, and can deliver significantly better ranking performance compared to state-of-the-art fair ranking methods.
Improving Position Encoding of Transformers for Multivariate Time Series Classification
Abstract
Transformers have demonstrated outstanding performance in many applications of deep learning. When applied to time series data, transformers require effective position encoding to capture the ordering of the time series data. The efficacy of position encoding in time series analysis is not well-studied and remains controversial, e.g., whether it is better to inject absolute position encoding or relative position encoding, or a combination of them. In order to clarify this, we first review existing absolute and relative position encoding methods when applied in time series classification. We then proposed a new absolute position encoding method dedicated to time series data called time Absolute Position Encoding (tAPE). Our new method incorporates the series length and input embedding dimension in absolute position encoding. Additionally, we propose computationally Efficient implementation of Relative Position Encoding (eRPE) to improve generalisability for time series. We then propose a novel multivariate time series classification (MTSC) model combining tAPE/eRPE and convolution-based input encoding named ConvTran to improve the position and data embedding of time series data. The proposed absolute and relative position encoding methods are simple and efficient. They can be easily integrated into transformer blocks and used for downstream tasks such as forecasting, extrinsic regression, and anomaly detection. Extensive experiments on 32 multivariate time-series datasets show that our model is significantly more accurate than state-of-the-art convolution and transformer-based models. Code and models are open-sourced at \url{https://github.com/Navidfoumani/ConvTran}.
CAILA: Concept-Aware Intra-Layer Adapters for Compositional Zero-Shot Learning
Authors: Zhaoheng Zheng, Haidong Zhu, Ram Nevatia
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Compositionality, the ability to combine existing concepts and generalize towards novel compositions, is a key functionality for intelligent entities. Here, we study the problem of Compositional Zero-Shot Learning (CZSL), which aims at recognizing novel attribute-object compositions. Recent approaches build their systems on top of large-scale Vision-Language Pre-trained (VLP) models, e.g. CLIP, and observe significant improvements. However, these methods treat CLIP as a black box and focus on pre- and post-CLIP operations. Here, we propose to dive deep into the architecture and insert adapters, a parameter-efficient technique proven to be effective among large language models, to each CLIP encoder layer. We further equip adapters with concept awareness so that concept-specific features of "object", "attribute" and "composition" can be extracted. We name our method CAILA, Concept-Aware Intra-Layer Adapters. Quantitative evaluations performed on three popular CZSL datasets, MIT-States, C-GQA, and UT-Zappos, reveal that CAILA achieves double-digit relative improvements against the current state-of-the-art on all benchmarks.
Sharpend Cosine Similarity based Neural Network for Hyperspectral Image Classification
Authors: Muhammad Ahmad
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Hyperspectral Image Classification (HSIC) is a difficult task due to high inter and intra-class similarity and variability, nested regions, and overlapping. 2D Convolutional Neural Networks (CNN) emerged as a viable network whereas, 3D CNNs are a better alternative due to accurate classification. However, 3D CNNs are highly computationally complex due to their volume and spectral dimensions. Moreover, down-sampling and hierarchical filtering (high frequency) i.e., texture features need to be smoothed during the forward pass which is crucial for accurate HSIC. Furthermore, CNN requires tons of tuning parameters which increases the training time. Therefore, to overcome the aforesaid issues, Sharpened Cosine Similarity (SCS) concept as an alternative to convolutions in a Neural Network for HSIC is introduced. SCS is exceptionally parameter efficient due to skipping the non-linear activation layers, normalization, and dropout after the SCS layer. Use of MaxAbsPool instead of MaxPool which selects the element with the highest magnitude of activity, even if it's negative. Experimental results on publicly available HSI datasets proved the performance of SCS as compared to the convolutions in Neural Networks.
Future-conditioned Unsupervised Pretraining for Decision Transformer
Authors: Zhihui Xie, Zichuan Lin, Deheng Ye, Qiang Fu, Wei Yang, Shuai Li
Abstract
Recent research in offline reinforcement learning (RL) has demonstrated that return-conditioned supervised learning is a powerful paradigm for decision-making problems. While promising, return conditioning is limited to training data labeled with rewards and therefore faces challenges in learning from unsupervised data. In this work, we aim to utilize generalized future conditioning to enable efficient unsupervised pretraining from reward-free and sub-optimal offline data. We propose Pretrained Decision Transformer (PDT), a conceptually simple approach for unsupervised RL pretraining. PDT leverages future trajectory information as a privileged context to predict actions during training. The ability to make decisions based on both present and future factors enhances PDT's capability for generalization. Besides, this feature can be easily incorporated into a return-conditioned framework for online finetuning, by assigning return values to possible futures and sampling future embeddings based on their respective values. Empirically, PDT outperforms or performs on par with its supervised pretraining counterpart, especially when dealing with sub-optimal data. Further analysis reveals that PDT can extract diverse behaviors from offline data and controllably sample high-return behaviors by online finetuning. Code is available at here.
Kaczmarz-Type Method for Solving Matrix Equation $AXB=C$
Abstract
In this paper, several row and column orthogonal projection methods are proposed for solving matrix equation $AXB=C$, where the matrix $A$ and $B$ are full rank or rank deficient and equation is consistent or not. These methods are iterative methods without matrix multiplication. It is theoretically proved these methods converge to the solution or least-squares solution of the matrix equation. Numerical results show that these methods are more efficient than iterative methods involving matrix multiplication for high-dimensional matrix.
PIP: Parse-Instructed Prefix for Syntactically Controlled Paraphrase Generation
Authors: Yixin Wan, Kuan-Hao Huang, Kai-Wei Chang
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
Syntactically controlled paraphrase generation requires language models to generate paraphrases for sentences according to specific syntactic structures. Existing fine-tuning methods for this task are costly as all the parameters of the model need to be updated during the training process. Inspired by recent studies on parameter-efficient learning, we propose Parse-Instructed Prefix (PIP), a novel adaptation of prefix-tuning to tune large pre-trained language models on syntactically controlled paraphrase generation task in a low-data setting with significantly less training cost. We introduce two methods to instruct a model's encoder prefix to capture syntax-related knowledge: direct initiation (PIP-Direct) and indirect optimization (PIP-Indirect). In contrast to traditional fine-tuning methods for this task, PIP is a compute-efficient alternative with 10 times less learnable parameters. Compared to existing prefix-tuning methods, PIP excels at capturing syntax control information, achieving significantly higher performance at the same level of learnable parameter count.
Merging control in mixed traffic with safety guarantees: a safe sequencing policy with optimal motion control
Authors: Ehsan Sabouni, H.M. Sabbir Ahmad, Christos G. Cassandras, Wenchao Li
Abstract
We address the problem of merging traffic from two roadways consisting of both Connected Autonomous Vehicles (CAVs) and Human Driven Vehicles (HDVs). Guaranteeing safe merging in such mixed traffic settings is challenging due to the unpredictability of possibly uncooperative HDVs. We develop a hierarchical controller where at each discrete time step first a coordinator determines the best possible Safe Sequence (SS) which can be realized without any knowledge of human driving behavior. Then, a lower-level decentralized motion controller for each CAV jointly minimizes travel time and energy over a prediction horizon, subject to hard safety constraints dependent on the given safe sequence. This is accomplished using a Model Predictive Controller (MPC) subject to constraints based on Control Barrier Functions (CBFs) which render it computationally efficient. Extensive simulation results are included showing that this hierarchical controller outperforms the commonly adopted Shortest Distance First (SDF) passing sequence over the full range of CAV penetration rates, while also providing safe merging guarantees.
Symmetric resonance based integrators and forest formulae
Authors: Yvonne Alama Bronsard, Yvain Bruned, Georg Maierhofer, Katharina Schratz
Subjects: Numerical Analysis (math.NA); Analysis of PDEs (math.AP); Rings and Algebras (math.RA)
Abstract
We introduce a unified framework of symmetric resonance based schemes which preserve central symmetries of the underlying PDE. We extend the resonance decorated trees approach introduced in arXiv:2005.01649 to a richer framework by exploring novel ways of iterating Duhamel's formula, capturing the dominant parts while interpolating the lower parts of the resonances in a symmetric manner. This gives a general class of new numerical schemes with more degrees of freedom than the original scheme from arXiv:2005.01649. To encapsulate the central structures we develop new forest formulae that contain the previous class of schemes and derive conditions on their coefficients in order to obtain symmetric schemes. These forest formulae echo the one used in Quantum Field Theory for renormalising Feynman diagrams and the one used for the renormalisation of singular SPDEs via the theory of Regularity Structures. These new algebraic tools not only provide a nice parametrisation of the previous resonance based integrators but also allow us to find new symmetric schemes with remarkable structure preservation properties even at very low regularity.
Parameter-Efficient Fine-Tuning without Introducing New Latency
Authors: Baohao Liao, Yan Meng, Christof Monz
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
Parameter-efficient fine-tuning (PEFT) of pre-trained language models has recently demonstrated remarkable achievements, effectively matching the performance of full fine-tuning while utilizing significantly fewer trainable parameters, and consequently addressing the storage and communication constraints. Nonetheless, various PEFT methods are limited by their inherent characteristics. In the case of sparse fine-tuning, which involves modifying only a small subset of the existing parameters, the selection of fine-tuned parameters is task- and domain-specific, making it unsuitable for federated learning. On the other hand, PEFT methods with adding new parameters typically introduce additional inference latency. In this paper, we demonstrate the feasibility of generating a sparse mask in a task-agnostic manner, wherein all downstream tasks share a common mask. Our approach, which relies solely on the magnitude information of pre-trained parameters, surpasses existing methodologies by a significant margin when evaluated on the GLUE benchmark. Additionally, we introduce a novel adapter technique that directly applies the adapter to pre-trained parameters instead of the hidden representation, thereby achieving identical inference speed to that of full fine-tuning. Through extensive experiments, our proposed method attains a new state-of-the-art outcome in terms of both performance and storage efficiency, storing only 0.03% parameters of full fine-tuning.
Leveraging Domain Knowledge for Inclusive and Bias-aware Humanitarian Response Entry Classification
Abstract
Accurate and rapid situation analysis during humanitarian crises is critical to delivering humanitarian aid efficiently and is fundamental to humanitarian imperatives and the Leave No One Behind (LNOB) principle. This data analysis can highly benefit from language processing systems, e.g., by classifying the text data according to a humanitarian ontology. However, approaching this by simply fine-tuning a generic large language model (LLM) involves considerable practical and ethical issues, particularly the lack of effectiveness on data-sparse and complex subdomains, and the encoding of societal biases and unwanted associations. In this work, we aim to provide an effective and ethically-aware system for humanitarian data analysis. We approach this by (1) introducing a novel architecture adjusted to the humanitarian analysis framework, (2) creating and releasing a novel humanitarian-specific LLM called HumBert, and (3) proposing a systematic way to measure and mitigate biases. Our experiments' results show the better performance of our approach on zero-shot and full-training settings in comparison with strong baseline models, while also revealing the existence of biases in the resulting LLMs. Utilizing a targeted counterfactual data augmentation approach, we significantly reduce these biases without compromising performance.
Towards a Common Understanding of Contributing Factors for Cross-Lingual Transfer in Multilingual Language Models: A Review
Authors: Fred Philippy, Siwen Guo, Shohreh Haddadan
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
In recent years, pre-trained Multilingual Language Models (MLLMs) have shown a strong ability to transfer knowledge across different languages. However, given that the aspiration for such an ability has not been explicitly incorporated in the design of the majority of MLLMs, it is challenging to obtain a unique and straightforward explanation for its emergence. In this review paper, we survey literature that investigates different factors contributing to the capacity of MLLMs to perform zero-shot cross-lingual transfer and subsequently outline and discuss these factors in detail. To enhance the structure of this review and to facilitate consolidation with future studies, we identify five categories of such factors. In addition to providing a summary of empirical evidence from past studies, we identify consensuses among studies with consistent findings and resolve conflicts among contradictory ones. Our work contextualizes and unifies existing research streams which aim at explaining the cross-lingual potential of MLLMs. This review provides, first, an aligned reference point for future research and, second, guidance for a better-informed and more efficient way of leveraging the cross-lingual capacity of MLLMs.
Uncertain Pose Estimation during Contact Tasks using Differentiable Contact Features
Abstract
For many robotic manipulation and contact tasks, it is crucial to accurately estimate uncertain object poses, for which certain geometry and sensor information are fused in some optimal fashion. Previous results for this problem primarily adopt sampling-based or end-to-end learning methods, which yet often suffer from the issues of efficiency and generalizability. In this paper, we propose a novel differentiable framework for this uncertain pose estimation during contact, so that it can be solved in an efficient and accurate manner with gradient-based solver. To achieve this, we introduce a new geometric definition that is highly adaptable and capable of providing differentiable contact features. Then we approach the problem from a bi-level perspective and utilize the gradient of these contact features along with differentiable optimization to efficiently solve for the uncertain pose. Several scenarios are implemented to demonstrate how the proposed framework can improve existing methods.
vFedSec: Efficient Secure Aggregation for Vertical Federated Learning via Secure Layer
Authors: Xinchi Qiu, Heng Pan, Wanru Zhao, Chenyang Ma, Pedro P.B. Gusmao, Nicholas D. Lane
Subjects: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
Abstract
Most work in privacy-preserving federated learning (FL) has been focusing on horizontally partitioned datasets where clients share the same sets of features and can train complete models independently. However, in many interesting problems, individual data points are scattered across different clients/organizations in a vertical setting. Solutions for this type of FL require the exchange of intermediate outputs and gradients between participants, posing a potential risk of privacy leakage when privacy and security concerns are not considered. In this work, we present vFedSec - a novel design with an innovative Secure Layer for training vertical FL securely and efficiently using state-of-the-art security modules in secure aggregation. We theoretically demonstrate that our method does not impact the training performance while protecting private data effectively. Empirically results also show its applicability with extensive experiments that our design can achieve the protection with negligible computation and communication overhead. Also, our method can obtain 9.1e2 ~ 3.8e4 speedup compared to widely-adopted homomorphic encryption (HE) method.
Joint Optimization of Triangle Mesh, Material, and Light from Neural Fields with Neural Radiance Cache
Abstract
Traditional inverse rendering techniques are based on textured meshes, which naturally adapts to modern graphics pipelines, but costly differentiable multi-bounce Monte Carlo (MC) ray tracing poses challenges for modeling global illumination. Recently, neural fields has demonstrated impressive reconstruction quality but falls short in modeling indirect illumination. In this paper, we introduce a simple yet efficient inverse rendering framework that combines the strengths of both methods. Specifically, given pre-trained neural field representing the scene, we can obtain an initial estimate of the signed distance field (SDF) and create a Neural Radiance Cache (NRC), an enhancement over the traditional radiance cache used in real-time rendering. By using the former to initialize differentiable marching tetrahedrons (DMTet) and the latter to model indirect illumination, we can compute the global illumination via single-bounce differentiable MC ray tracing and jointly optimize the geometry, material, and light through back propagation. Experiments demonstrate that, compared to previous methods, our approach effectively prevents indirect illumination effects from being baked into materials, thus obtaining the high-quality reconstruction of triangle mesh, Physically-Based (PBR) materials, and High Dynamic Range (HDR) light probe.
Domain Aligned Prefix Averaging for Domain Generalization in Abstractive Summarization
Abstract
Domain generalization is hitherto an underexplored area applied in abstractive summarization. Moreover, most existing works on domain generalization have sophisticated training algorithms. In this paper, we propose a lightweight, weight averaging based, Domain Aligned Prefix Averaging approach to domain generalization for abstractive summarization. Given a number of source domains, our method first trains a prefix for each one of them. These source prefixes generate summaries for a small number of target domain documents. The similarity of the generated summaries to their corresponding documents is used for calculating weights required to average source prefixes. In DAPA, prefix tuning allows for lightweight finetuning, and weight averaging allows for the computationally efficient addition of new source domains. When evaluated on four diverse summarization domains, DAPA shows comparable or better performance against the baselines, demonstrating the effectiveness of its prefix averaging scheme.
Randomized Positional Encodings Boost Length Generalization of Transformers
Authors: Anian Ruoss, Grégoire Delétang, Tim Genewein, Jordi Grau-Moya, Róbert Csordás, Mehdi Bennani, Shane Legg, Joel Veness
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (stat.ML)
Abstract
Transformers have impressive generalization capabilities on tasks with a fixed context length. However, they fail to generalize to sequences of arbitrary length, even for seemingly simple tasks such as duplicating a string. Moreover, simply training on longer sequences is inefficient due to the quadratic computation complexity of the global attention mechanism. In this work, we demonstrate that this failure mode is linked to positional encodings being out-of-distribution for longer sequences (even for relative encodings) and introduce a novel family of positional encodings that can overcome this problem. Concretely, our randomized positional encoding scheme simulates the positions of longer sequences and randomly selects an ordered subset to fit the sequence's length. Our large-scale empirical evaluation of 6000 models across 15 algorithmic reasoning tasks shows that our method allows Transformers to generalize to sequences of unseen length (increasing test accuracy by 12.0% on average).
Green Runner: A tool for efficient model selection from model repositories
Authors: Jai Kannan, Scott Barnett, Anj Simmons, Taylan Selvi, Luis Cruz
Abstract
Deep learning models have become essential in software engineering, enabling intelligent features like image captioning and document generation. However, their popularity raises concerns about environmental impact and inefficient model selection. This paper introduces GreenRunnerGPT, a novel tool for efficiently selecting deep learning models based on specific use cases. It employs a large language model to suggest weights for quality indicators, optimizing resource utilization. The tool utilizes a multi-armed bandit framework to evaluate models against target datasets, considering tradeoffs. We demonstrate that GreenRunnerGPT is able to identify a model suited to a target use case without wasteful computations that would occur under a brute-force approach to model selection.
Automation of Trimming Die Design Inspection by Zigzag Process Between AI and CAD Domains
Authors: Jinsub Lee, Tae-Hyun Kim, Sang-Hwan Jeon, Sung-Hyun Park, Sang-Hi Kim, Eun-Ho Lee, Jee-Hyong Lee
Abstract
Quality control in the manufacturing industry has improved with the use of artificial intelligence (AI). However, the manual inspection of trimming die designs, which is time-consuming and prone to errors, is still done by engineers. This study introduces an automatic design inspection system for automobile trimming dies by integrating AI modules and computer-aided design (CAD) software. The AI modules replace engineers' judgment, and the CAD software carries out operations requested by the AI modules. The inspection process involves a zigzag interaction between the AI modules and CAD software, enabling one-click operation without expert intervention. The AI modules are CAD-independent and data-efficient, making them adaptable to other CAD software. They achieve high performance even with limited training data, with an average length measurement error of only 2.4%. The inspection time is reduced to approximately one-fifth of the time required for manual inspection by experts.
Modelling, Analysis and Control of OmniMorph:an Omnidirectional Morphing Multi-rotor UAV
Authors: Youssef Aboudorra, Chiara Gabellieri, Quentin Sablé, Antonio Franchi
Subjects: Robotics (cs.RO); Systems and Control (eess.SY)
Abstract
We present the design, modelling, and control of a novel morphing multi-rotor Unmanned Aerial Vehicle (UAV) that we call the OmniMorph. The morphing ability allows the platform to switch between different configurations to achieve the required task. The uni-directional thrust (UDT) configuration can be used for energy-efficient navigation, while fully-actuated (FA) and omnidirectional (OD) configurations can be used for full pose tracking and make the platform assume any orientation while compensating the gravity. The platform is equipped with eight bi-directional propellers that are actively tilted in a synchronized fashion using only one additional degree of actuation.
Distributional Reinforcement Learning with Dual Expectile-Quantile Regression
Authors: Sami Jullien, Romain Deffayet, Jean-Michel Renders, Paul Groth, Maarten de Rijke
Abstract
Successful applications of distributional reinforcement learning with quantile regression prompt a natural question: can we use other statistics to represent the distribution of returns? In particular, expectile regression is known to be more efficient than quantile regression for approximating distributions, especially on extreme values, and by providing a straightforward estimator of the mean it is a natural candidate for reinforcement learning. Prior work has answered this question positively in the case of expectiles, with the major caveat that expensive computations must be performed to ensure convergence. In this work, we propose a dual expectile-quantile approach which solves the shortcomings of previous work while leveraging the complementary properties of expectiles and quantiles. Our method outperforms both quantile-based and expectile-based baselines on the MuJoCo continuous control benchmark.
Peeking inside Sparse Neural Networks using Multi-Partite Graph Representations
Authors: Elia Cunegatti, Doina Bucur, Giovanni Iacca
Abstract
Modern Deep Neural Networks (DNNs) have achieved very high performance at the expense of computational resources. To decrease the computational burden, several techniques have proposed to extract, from a given DNN, efficient subnetworks which are able to preserve performance while reducing the number of network parameters. The literature provides a broad set of techniques to discover such subnetworks, but few works have studied the peculiar topologies of such pruned architectures. In this paper, we propose a novel \emph{unrolled input-aware} bipartite Graph Encoding (GE) that is able to generate, for each layer in an either sparse or dense neural network, its corresponding graph representation based on its relation with the input data. We also extend it into a multipartite GE, to capture the relation between layers. Then, we leverage on topological properties to study the difference between the existing pruning algorithms and algorithm categories, as well as the relation between topologies and performance.
Feature Adaptation for Sparse Linear Regression
Authors: Jonathan Kelner, Frederic Koehler, Raghu Meka, Dhruv Rohatgi
Subjects: Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG); Statistics Theory (math.ST); Machine Learning (stat.ML)
Abstract
Sparse linear regression is a central problem in high-dimensional statistics. We study the correlated random design setting, where the covariates are drawn from a multivariate Gaussian $N(0,\Sigma)$, and we seek an estimator with small excess risk. If the true signal is $t$-sparse, information-theoretically, it is possible to achieve strong recovery guarantees with only $O(t\log n)$ samples. However, computationally efficient algorithms have sample complexity linear in (some variant of) the condition number of $\Sigma$. Classical algorithms such as the Lasso can require significantly more samples than necessary even if there is only a single sparse approximate dependency among the covariates. We provide a polynomial-time algorithm that, given $\Sigma$, automatically adapts the Lasso to tolerate a small number of approximate dependencies. In particular, we achieve near-optimal sample complexity for constant sparsity and if $\Sigma$ has few ``outlier'' eigenvalues. Our algorithm fits into a broader framework of feature adaptation for sparse linear regression with ill-conditioned covariates. With this framework, we additionally provide the first polynomial-factor improvement over brute-force search for constant sparsity $t$ and arbitrary covariance $\Sigma$.
Abstract
Despite the rich existing literature about minimax optimization in continuous settings, only very partial results of this kind have been obtained for combinatorial settings. In this paper, we fill this gap by providing a characterization of submodular minimax optimization, the problem of finding a set (for either the min or the max player) that is effective against every possible response. We show when and under what conditions we can find such sets. We also demonstrate how minimax submodular optimization provides robust solutions for downstream machine learning applications such as (i) efficient prompt engineering for question answering, (ii) prompt engineering for dialog state tracking, (iii) identifying robust waiting locations for ride-sharing, (iv) ride-share difficulty kernelization, and (v) finding adversarial images. Our experiments demonstrate that our proposed algorithms consistently outperform other baselines.
A Neural State-Space Model Approach to Efficient Speech Separation
Authors: Chen Chen, Chao-Han Huck Yang, Kai Li, Yuchen Hu, Pin-Jui Ku, Eng Siong Chng
Subjects: Sound (cs.SD); Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
Abstract
In this work, we introduce S4M, a new efficient speech separation framework based on neural state-space models (SSM). Motivated by linear time-invariant systems for sequence modeling, our SSM-based approach can efficiently model input signals into a format of linear ordinary differential equations (ODEs) for representation learning. To extend the SSM technique into speech separation tasks, we first decompose the input mixture into multi-scale representations with different resolutions. This mechanism enables S4M to learn globally coherent separation and reconstruction. The experimental results show that S4M performs comparably to other separation backbones in terms of SI-SDRi, while having a much lower model complexity with significantly fewer trainable parameters. In addition, our S4M-tiny model (1.8M parameters) even surpasses attention-based Sepformer (26.0M parameters) in noisy conditions with only 9.2 of multiply-accumulate operation (MACs).
DiffusionNAG: Task-guided Neural Architecture Generation with Diffusion Models
Authors: Sohyun An, Hayeon Lee, Jaehyeong Jo, Seanie Lee, Sung Ju Hwang
Abstract
Neural Architecture Search (NAS) has emerged as a powerful technique for automating neural architecture design. However, existing NAS methods either require an excessive amount of time for repetitive training or sampling of many task-irrelevant architectures. Moreover, they lack generalization across different tasks and usually require searching for optimal architectures for each task from scratch without reusing the knowledge from the previous NAS tasks. To tackle such limitations of existing NAS methods, we propose a novel transferable task-guided Neural Architecture Generation (NAG) framework based on diffusion models, dubbed DiffusionNAG. With the guidance of a surrogate model, such as a performance predictor for a given task, our DiffusionNAG can generate task-optimal architectures for diverse tasks, including unseen tasks. DiffusionNAG is highly efficient as it generates task-optimal neural architectures by leveraging the prior knowledge obtained from the previous tasks and neural architecture distribution. Furthermore, we introduce a score network to ensure the generation of valid architectures represented as directed acyclic graphs, unlike existing graph generative models that focus on generating undirected graphs. Extensive experiments demonstrate that DiffusionNAG significantly outperforms the state-of-the-art transferable NAG model in architecture generation quality, as well as previous NAS methods on four computer vision datasets with largely reduced computational cost.
Meta-prediction Model for Distillation-Aware NAS on Unseen Datasets
Authors: Hayeon Lee, Sohyun An, Minseon Kim, Sung Ju Hwang
Abstract
Distillation-aware Neural Architecture Search (DaNAS) aims to search for an optimal student architecture that obtains the best performance and/or efficiency when distilling the knowledge from a given teacher model. Previous DaNAS methods have mostly tackled the search for the neural architecture for fixed datasets and the teacher, which are not generalized well on a new task consisting of an unseen dataset and an unseen teacher, thus need to perform a costly search for any new combination of the datasets and the teachers. For standard NAS tasks without KD, meta-learning-based computationally efficient NAS methods have been proposed, which learn the generalized search process over multiple tasks (datasets) and transfer the knowledge obtained over those tasks to a new task. However, since they assume learning from scratch without KD from a teacher, they might not be ideal for DaNAS scenarios. To eliminate the excessive computational cost of DaNAS methods and the sub-optimality of rapid NAS methods, we propose a distillation-aware meta accuracy prediction model, DaSS (Distillation-aware Student Search), which can predict a given architecture's final performances on a dataset when performing KD with a given teacher, without having actually to train it on the target task. The experimental results demonstrate that our proposed meta-prediction model successfully generalizes to multiple unseen datasets for DaNAS tasks, largely outperforming existing meta-NAS methods and rapid NAS baselines. Code is available at https://github.com/CownowAn/DaSS
CUQIpy -- Part I: computational uncertainty quantification for inverse problems in Python
Authors: Nicolai A B Riis, Amal M A Alghamdi, Felipe Uribe, Silja L Christensen, Babak M Afkham, Per Christian Hansen, Jakob S Jørgensen
Abstract
This paper introduces CUQIpy, a versatile open-source Python package for computational uncertainty quantification (UQ) in inverse problems, presented as Part I of a two-part series. CUQIpy employs a Bayesian framework, integrating prior knowledge with observed data to produce posterior probability distributions that characterize the uncertainty in computed solutions to inverse problems. The package offers a high-level modeling framework with concise syntax, allowing users to easily specify their inverse problems, prior information, and statistical assumptions. CUQIpy supports a range of efficient sampling strategies and is designed to handle large-scale problems. Notably, the automatic sampler selection feature analyzes the problem structure and chooses a suitable sampler without user intervention, streamlining the process. With a selection of probability distributions, test problems, computational methods, and visualization tools, CUQIpy serves as a powerful, flexible, and adaptable tool for UQ in a wide selection of inverse problems. Part II of the series focuses on the use of CUQIpy for UQ in inverse problems with partial differential equations (PDEs).
Implementation-Efficient Finite Alphabet Decoding of Polar Codes
Authors: Philipp Mohr, Syed Aizaz Ali Shah, Gerhard Bauch
Abstract
An implementation-efficient finite alphabet decoder for polar codes relying on coarsely quantized messages and low-complexity operations is proposed. Typically, finite alphabet decoding performs concatenated compression operations on the received channel messages to aggregate compact reliability information for error correction. These compression operations or mappings can be considered as lookup tables. For polar codes, the finite alphabet decoder design boils down to constructing lookup tables for the upper and lower branches of the building blocks within the code structure. A key challenge is to realize a hardware-friendly implementation of the lookup tables. This work uses the min-sum implementation for the upper branch lookup table and, as a novelty, a computational domain implementation for the lower branch lookup table. The computational domain approach drastically reduces the number of implementation parameters. Furthermore, a restriction to uniform quantization in the lower branch allows a very hardware-friendly compression via clipping and bit-shifting. Its behavior is close to the optimal non-uniform quantization, whose implementation would require multiple high-resolution threshold comparisons. Simulation results confirm excellent performance for the developed decoder. Unlike conventional fixed-point decoders, the proposed method involves an offline design that explicitly maximizes the preserved mutual information under coarse quantization.
Training Socially Aligned Language Models in Simulated Human Society
Authors: Ruibo Liu, Ruixin Yang, Chenyan Jia, Ge Zhang, Denny Zhou, Andrew M. Dai, Diyi Yang, Soroush Vosoughi
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
Abstract
Social alignment in AI systems aims to ensure that these models behave according to established societal values. However, unlike humans, who derive consensus on value judgments through social interaction, current language models (LMs) are trained to rigidly replicate their training corpus in isolation, leading to subpar generalization in unfamiliar scenarios and vulnerability to adversarial attacks. This work presents a novel training paradigm that permits LMs to learn from simulated social interactions. In comparison to existing methodologies, our approach is considerably more scalable and efficient, demonstrating superior performance in alignment benchmarks and human evaluations. This paradigm shift in the training of LMs brings us a step closer to developing AI systems that can robustly and accurately reflect societal norms and values.
TranSFormer: Slow-Fast Transformer for Machine Translation
Authors: Bei Li, Yi Jing, Xu Tan, Zhen Xing, Tong Xiao, Jingbo Zhu
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
Learning multiscale Transformer models has been evidenced as a viable approach to augmenting machine translation systems. Prior research has primarily focused on treating subwords as basic units in developing such systems. However, the incorporation of fine-grained character-level features into multiscale Transformer has not yet been explored. In this work, we present a \textbf{S}low-\textbf{F}ast two-stream learning model, referred to as Tran\textbf{SF}ormer, which utilizes a slow'' branch to deal with subword sequences and afast'' branch to deal with longer character sequences. This model is efficient since the fast branch is very lightweight by reducing the model width, and yet provides useful fine-grained features for the slow branch. Our TranSFormer shows consistent BLEU improvements (larger than 1 BLEU point) on several machine translation benchmarks.
A Tale of Two Approximations: Tightening Over-Approximation for DNN Robustness Verification via Under-Approximation
Authors: Zhiyi Xue, Si Liu, Zhaodi Zhang, Yiting Wu, Min Zhang
Abstract
The robustness of deep neural networks (DNNs) is crucial to the hosting system's reliability and security. Formal verification has been demonstrated to be effective in providing provable robustness guarantees. To improve its scalability, over-approximating the non-linear activation functions in DNNs by linear constraints has been widely adopted, which transforms the verification problem into an efficiently solvable linear programming problem. Many efforts have been dedicated to defining the so-called tightest approximations to reduce overestimation imposed by over-approximation. In this paper, we study existing approaches and identify a dominant factor in defining tight approximation, namely the approximation domain of the activation function. We find out that tight approximations defined on approximation domains may not be as tight as the ones on their actual domains, yet existing approaches all rely only on approximation domains. Based on this observation, we propose a novel dual-approximation approach to tighten over-approximations, leveraging an activation function's underestimated domain to define tight approximation bounds. We implement our approach with two complementary algorithms based respectively on Monte Carlo simulation and gradient descent into a tool called DualApp. We assess it on a comprehensive benchmark of DNNs with different architectures. Our experimental results show that DualApp significantly outperforms the state-of-the-art approaches with 100% - 1000% improvement on the verified robustness ratio and 10.64% on average (up to 66.53%) on the certified lower bound.
Aggregating Capacity in FL through Successive Layer Training for Computationally-Constrained Devices
Abstract
Federated learning (FL) is usually performed on resource-constrained edge devices, e.g., with limited memory for the computation. If the required memory to train a model exceeds this limit, the device will be excluded from the training. This can lead to a lower accuracy as valuable data and computation resources are excluded from training, also causing bias and unfairness. The FL training process should be adjusted to such constraints. The state-of-the-art techniques propose training subsets of the FL model at constrained devices, reducing their resource requirements for training. But these techniques largely limit the co-adaptation among parameters of the model and are highly inefficient, as we show: it is actually better to train a smaller (less accurate) model by the system where all the devices can train the model end-to-end, than applying such techniques. We propose a new method that enables successive freezing and training of the parameters of the FL model at devices, reducing the training's resource requirements at the devices, while still allowing enough co-adaptation between parameters. We show through extensive experimental evaluation that our technique greatly improves the accuracy of the trained model (by 52.4 p.p.) compared with the state of the art, efficiently aggregating the computation capacity available on distributed devices.
Let the Flows Tell: Solving Graph Combinatorial Optimization Problems with GFlowNets
Authors: Dinghuai Zhang, Hanjun Dai, Nikolay Malkin, Aaron Courville, Yoshua Bengio, Ling Pan
Abstract
Combinatorial optimization (CO) problems are often NP-hard and thus out of reach for exact algorithms, making them a tempting domain to apply machine learning methods. The highly structured constraints in these problems can hinder either optimization or sampling directly in the solution space. On the other hand, GFlowNets have recently emerged as a powerful machinery to efficiently sample from composite unnormalized densities sequentially and have the potential to amortize such solution-searching processes in CO, as well as generate diverse solution candidates. In this paper, we design Markov decision processes (MDPs) for different combinatorial problems and propose to train conditional GFlowNets to sample from the solution space. Efficient training techniques are also developed to benefit long-range credit assignment. Through extensive experiments on a variety of different CO tasks with synthetic and realistic data, we demonstrate that GFlowNet policies can efficiently find high-quality solutions.
Formal Modelling for Multi-Robot Systems Under Uncertainty
Authors: Charlie Street, Masoumeh Mansouri, Bruno Lacerda
Subjects: Robotics (cs.RO); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
Abstract
Purpose of Review: To effectively synthesise and analyse multi-robot behaviour, we require formal task-level models which accurately capture multi-robot execution. In this paper, we review modelling formalisms for multi-robot systems under uncertainty, and discuss how they can be used for planning, reinforcement learning, model checking, and simulation. Recent Findings: Recent work has investigated models which more accurately capture multi-robot execution by considering different forms of uncertainty, such as temporal uncertainty and partial observability, and modelling the effects of robot interactions on action execution. Other strands of work have presented approaches for reducing the size of multi-robot models to admit more efficient solution methods. This can be achieved by decoupling the robots under independence assumptions, or reasoning over higher level macro actions. Summary: Existing multi-robot models demonstrate a trade off between accurately capturing robot dependencies and uncertainty, and being small enough to tractably solve real world problems. Therefore, future research should exploit realistic assumptions over multi-robot behaviour to develop smaller models which retain accurate representations of uncertainty and robot interactions; and exploit the structure of multi-robot problems, such as factored state spaces, to develop scalable solution methods.
Diable: Efficient Dialogue State Tracking as Operations on Tables
Abstract
Sequence-to-sequence state-of-the-art systems for dialogue state tracking (DST) use the full dialogue history as input, represent the current state as a list with all the slots, and generate the entire state from scratch at each dialogue turn. This approach is inefficient, especially when the number of slots is large and the conversation is long. In this paper, we propose Diable, a new task formalisation that simplifies the design and implementation of efficient DST systems and allows one to easily plug and play large language models. We represent the dialogue state as a table and formalise DST as a table manipulation task. At each turn, the system updates the previous state by generating table operations based on the dialogue context. Extensive experimentation on the MultiWoz datasets demonstrates that Diable (i) outperforms strong efficient DST baselines, (ii) is 2.4x more time efficient than current state-of-the-art methods while retaining competitive Joint Goal Accuracy, and (iii) is robust to noisy data annotations due to the table operations approach.
GLOBE-CE: A Translation-Based Approach for Global Counterfactual Explanations
Authors: Dan Ley, Saumitra Mishra, Daniele Magazzeni
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (stat.ML)
Abstract
Counterfactual explanations have been widely studied in explainability, with a range of application dependent methods prominent in fairness, recourse and model understanding. The major shortcoming associated with these methods, however, is their inability to provide explanations beyond the local or instance-level. While many works touch upon the notion of a global explanation, typically suggesting to aggregate masses of local explanations in the hope of ascertaining global properties, few provide frameworks that are both reliable and computationally tractable. Meanwhile, practitioners are requesting more efficient and interactive explainability tools. We take this opportunity to propose Global & Efficient Counterfactual Explanations (GLOBE-CE), a flexible framework that tackles the reliability and scalability issues associated with current state-of-the-art, particularly on higher dimensional datasets and in the presence of continuous features. Furthermore, we provide a unique mathematical analysis of categorical feature translations, utilising it in our method. Experimental evaluation with publicly available datasets and user studies demonstrate that GLOBE-CE performs significantly better than the current state-of-the-art across multiple metrics (e.g., speed, reliability).
Joint Antenna Selection and Beamforming for Massive MIMO-enabled Over-the-Air Federated Learning
Authors: Saba Asaad, Hina Tabassum, Chongjun Ouyang, Ping Wang
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Abstract
Over-the-air federated learning (OTA-FL) is an emerging technique to reduce the computation and communication overload at the PS caused by the orthogonal transmissions of the model updates in conventional federated learning (FL). This reduction is achieved at the expense of introducing aggregation error that can be efficiently suppressed by means of receive beamforming via large array-antennas. This paper studies OTA-FL in massive multiple-input multiple-output (MIMO) systems by considering a realistic scenario in which the edge server, despite its large antenna array, is restricted in the number of radio frequency (RF)-chains. For this setting, the beamforming for over-the-air model aggregation needs to be addressed jointly with antenna selection. This leads to an NP-hard problem due to the combinatorial nature of the optimization. We tackle this problem via two different approaches. In the first approach, we use the penalty dual decomposition (PDD) technique to develop a two-tier algorithm for joint antenna selection and beamforming. The second approach interprets the antenna selection task as a sparse recovery problem and develops two iterative joint algorithms based on the Lasso and fast iterative soft-thresholding methods. Convergence and complexity analysis is presented for all the schemes. The numerical investigations depict that the algorithms based on the sparse recovery techniques outperform the PDD-based algorithm, when the number of RF-chains at the edge server is much smaller than its array size. However, as the number of RF-chains increases, the PDD approach starts to be superior. Our simulations further depict that learning performance with all the antennas being active at the PS can be closely tracked by selecting less than 20% of the antennas at the PS.
Complete Multiparty Session Type Projection with Automata
Authors: Elaine Li, Felix Stutz, Thomas Wies, Damien Zufferey
Subjects: Formal Languages and Automata Theory (cs.FL); Distributed, Parallel, and Cluster Computing (cs.DC); Programming Languages (cs.PL)
Abstract
Multiparty session types (MSTs) are a type-based approach to verifying communication protocols. Central to MSTs is a projection operator: a partial function that maps protocols represented as global types to correct-by-construction implementations for each participant, represented as a communicating state machine. Existing projection operators are syntactic in nature, and trade efficiency for completeness. We present the first projection operator that is sound, complete, and efficient. Our projection separates synthesis from checking implementability. For synthesis, we use a simple automata-theoretic construction; for checking implementability, we present succinct conditions that summarize insights into the property of implementability. We use these conditions to show that MST implementability is PSPACE-complete. This improves upon a previous decision procedure that is in EXPSPACE and applies to a smaller class of MSTs. We demonstrate the effectiveness of our approach using a prototype implementation, which handles global types not supported by previous work without sacrificing performance.
Communication-Efficient Reinforcement Learning in Swarm Robotic Networks for Maze Exploration
Abstract
Smooth coordination within a swarm robotic system is essential for the effective execution of collective robot missions. Having efficient communication is key to the successful coordination of swarm robots. This paper proposes a new communication-efficient decentralized cooperative reinforcement learning algorithm for coordinating swarm robots. It is made efficient by hierarchically building on the use of local information exchanges. We consider a case study application of maze solving through cooperation among a group of robots, where the time and costs are minimized while avoiding inter-robot collisions and path overlaps during exploration. With a solid theoretical basis, we extensively analyze the algorithm with realistic CORE network simulations and evaluate it against state-of-the-art solutions in terms of maze coverage percentage and efficiency under communication-degraded environments. The results demonstrate significantly higher coverage accuracy and efficiency while reducing costs and overlaps even in high packet loss and low communication range scenarios.
Keyword: faster
WeiAvg: Federated Learning Model Aggregation Promoting Data Diversity
Authors: Fan Dong, Ali Abbasi, Steve Drew, Henry Leung, Xin Wang, Jiayu Zhou
Abstract
Federated learning provides a promising privacy-preserving way for utilizing large-scale private edge data from massive Internet-of-Things (IoT) devices. While existing research extensively studied optimizing the learning process, computing efficiency, and communication overhead, one important and often overlooked aspect is that participants contribute predictive knowledge from their data, impacting the quality of the federated models learned. While FedAvg treats each client equally and assigns weight solely based on the number of samples, the diversity of samples on each client could greatly affect the local update performance and the final aggregated model. In this paper, we propose a novel approach to address this issue by introducing a Weighted Averaging (WeiAvg) framework that emphasizes updates from high-diversity clients and diminishes the influence of those from low-diversity clients. Specifically, we introduced a projection-based approximation method to estimate the diversity of client data, instead of the computation of an entropy. We use the approximation because the locally computed entropy may not be transmitted due to excess privacy risk. Extensive experimental results show that WeiAvg converges faster and achieves higher accuracy than the original FedAvg algorithm and FedProx.
Abstract
Circuit representation learning aims to obtain neural representations of circuit elements and has emerged as a promising research direction that can be applied to various EDA and logic reasoning tasks. Existing solutions, such as DeepGate, have the potential to embed both circuit structural information and functional behavior. However, their capabilities are limited due to weak supervision or flawed model design, resulting in unsatisfactory performance in downstream tasks. In this paper, we introduce DeepGate2, a novel functionality-aware learning framework that significantly improves upon the original DeepGate solution in terms of both learning effectiveness and efficiency. Our approach involves using pairwise truth table differences between sampled logic gates as training supervision, along with a well-designed and scalable loss function that explicitly considers circuit functionality. Additionally, we consider inherent circuit characteristics and design an efficient one-round graph neural network (GNN), resulting in an order of magnitude faster learning speed than the original DeepGate solution. Experimental results demonstrate significant improvements in two practical downstream tasks: logic synthesis and Boolean satisfiability solving. The code is available at https://github.com/cure-lab/DeepGate2
Scan and Snap: Understanding Training Dynamics and Token Composition in 1-layer Transformer
Authors: Yuandong Tian, Yiping Wang, Beidi Chen, Simon Du
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
Transformer architecture has shown impressive performance in multiple research domains and has become the backbone of many neural network models. However, there is limited understanding on how it works. In particular, with a simple predictive loss, how the representation emerges from the gradient \emph{training dynamics} remains a mystery. In this paper, for 1-layer transformer with one self-attention layer plus one decoder layer, we analyze its SGD training dynamics for the task of next token prediction in a mathematically rigorous manner. We open the black box of the dynamic process of how the self-attention layer combines input tokens, and reveal the nature of underlying inductive bias. More specifically, with the assumption (a) no positional encoding, (b) long input sequence, and (c) the decoder layer learns faster than the self-attention layer, we prove that self-attention acts as a \emph{discriminative scanning algorithm}: starting from uniform attention, it gradually attends more to distinct key tokens for a specific next token to be predicted, and pays less attention to common key tokens that occur across different next tokens. Among distinct tokens, it progressively drops attention weights, following the order of low to high co-occurrence between the key and the query token in the training set. Interestingly, this procedure does not lead to winner-takes-all, but decelerates due to a \emph{phase transition} that is controllable by the learning rates of the two layers, leaving (almost) fixed token combination. We verify this \textbf{\emph{scan and snap}} dynamics on synthetic and real-world data (WikiText).
Metaheuristic planner for cooperative multi-agent wall construction with UAVs
Authors: Basel Elkhapery, Robert Pěnička, Michal Němec, Mohsin Siddiqui
Abstract
This paper introduces a wall construction planner for Unmanned Aerial Vehicles (UAVs), which uses a Greedy Randomized Adaptive Search Procedure (GRASP) metaheuristic to generate near-time-optimal building plans for even large walls within seconds. This approach addresses one of the most time-consuming and labor-intensive tasks, while also minimizing workers' safety risks. To achieve this, the wall-building problem is modeled as a variant of the Team Orienteering Problem and is formulated as Mixed-Integer Linear Programming (MILP), with added precedence and concurrence constraints that ensure bricks are built in the correct order and without collision between cooperating agents. The GRASP planner is validated in a realistic simulation and demonstrated to find solutions with similar quality as the optimal MILP, but much faster. Moreover, it outperforms all other state-of-the-art planning approaches in the majority of test cases. This paper presents a significant advancement in the field of automated wall construction, demonstrating the potential of UAVs and optimization algorithms in improving the efficiency and safety of construction projects.
Accelerating Value Iteration with Anchoring
Authors: Jongmin Lee, Ernest K. Ryu
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)
Abstract
Value Iteration (VI) is foundational to the theory and practice of modern reinforcement learning, and it is known to converge at a $\mathcal{O}(\gamma^k)$-rate, where $\gamma$ is the discount factor. Surprisingly, however, the optimal rate for the VI setup was not known, and finding a general acceleration mechanism has been an open problem. In this paper, we present the first accelerated VI for both the Bellman consistency and optimality operators. Our method, called Anc-VI, is based on an \emph{anchoring} mechanism (distinct from Nesterov's acceleration), and it reduces the Bellman error faster than standard VI. In particular, Anc-VI exhibits a $\mathcal{O}(1/k)$-rate for $\gamma\approx 1$ or even $\gamma=1$, while standard VI has rate $\mathcal{O}(1)$ for $\gamma\ge 1-1/k$, where $k$ is the iteration count. We also provide a complexity lower bound matching the upper bound up to a constant factor of $4$, thereby establishing optimality of the accelerated rate of Anc-VI. Finally, we show that the anchoring mechanism provides the same benefit in the approximate VI and Gauss--Seidel VI setups as well.
A Slingshot Approach to Learning in Monotone Games
Abstract
In this paper, we address the problem of computing equilibria in monotone games. The traditional Follow the Regularized Leader algorithms fail to converge to an equilibrium even in two-player zero-sum games. Although optimistic versions of these algorithms have been proposed with last-iterate convergence guarantees, they require noiseless gradient feedback. To overcome this limitation, we present a novel framework that achieves last-iterate convergence even in the presence of noise. Our key idea involves perturbing or regularizing the payoffs or utilities of the games. This perturbation serves to pull the current strategy to an anchored strategy, which we refer to as a {\it slingshot} strategy. First, we establish the convergence rates of our framework to a stationary point near an equilibrium, regardless of the presence or absence of noise. Next, we introduce an approach to periodically update the slingshot strategy with the current strategy. We interpret this approach as a proximal point method and demonstrate its last-iterate convergence. Our framework is comprehensive, incorporating existing payoff-regularized algorithms and enabling the development of new algorithms with last-iterate convergence properties. Finally, we show that our algorithms, based on this framework, empirically exhibit faster convergence.
Negative-prompt Inversion: Fast Image Inversion for Editing with Text-guided Diffusion Models
Abstract
In image editing employing diffusion models, it is crucial to preserve the reconstruction quality of the original image while changing its style. Although existing methods ensure reconstruction quality through optimization, a drawback of these is the significant amount of time required for optimization. In this paper, we propose negative-prompt inversion, a method capable of achieving equivalent reconstruction solely through forward propagation without optimization, thereby enabling much faster editing processes. We experimentally demonstrate that the reconstruction quality of our method is comparable to that of existing methods, allowing for inversion at a resolution of 512 pixels and with 50 sampling steps within approximately 5 seconds, which is more than 30 times faster than null-text inversion. Reduction of the computation time by the proposed method further allows us to use a larger number of sampling steps in diffusion models to improve the reconstruction quality with a moderate increase in computation time.
Channel and Gradient-Importance Aware Device Scheduling for Over-the-Air Federated Learning
Authors: Yuchang Sun, Zehong lin, Yuyi Mao, Shi Jin, Jun Zhang
Subjects: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract
Federated learning (FL) is a popular privacy-preserving distributed training scheme, where multiple devices collaborate to train machine learning models by uploading local model updates. To improve communication efficiency, over-the-air computation (AirComp) has been applied to FL, which leverages analog modulation to harness the superposition property of radio waves such that numerous devices can upload their model updates concurrently for aggregation. However, the uplink channel noise incurs considerable model aggregation distortion, which is critically determined by the device scheduling and compromises the learned model performance. In this paper, we propose a probabilistic device scheduling framework for over-the-air FL, named PO-FL, to mitigate the negative impact of channel noise, where each device is scheduled according to a certain probability and its model update is reweighted using this probability in aggregation. We prove the unbiasedness of this aggregation scheme and demonstrate the convergence of PO-FL on both convex and non-convex loss functions. Our convergence bounds unveil that the device scheduling affects the learning performance through the communication distortion and global update variance. Based on the convergence analysis, we further develop a channel and gradient-importance aware algorithm to optimize the device scheduling probabilities in PO-FL. Extensive simulation results show that the proposed PO-FL framework with channel and gradient-importance awareness achieves faster convergence and produces better models than baseline methods.
Can You Solve Closest String Faster than Exhaustive Search?
Authors: Amir Abboud, Nick Fischer, Elazar Goldenberg, Karthik C. S., Ron Safier
Subjects: Computational Complexity (cs.CC); Computational Geometry (cs.CG); Data Structures and Algorithms (cs.DS); Machine Learning (cs.LG)
Abstract
We study the fundamental problem of finding the best string to represent a given set, in the form of the Closest String problem: Given a set $X \subseteq \Sigma^d$ of $n$ strings, find the string $x^$ minimizing the radius of the smallest Hamming ball around $x^$ that encloses all the strings in $X$. In this paper, we investigate whether the Closest String problem admits algorithms that are faster than the trivial exhaustive search algorithm. We obtain the following results for the two natural versions of the problem: $\bullet$ In the continuous Closest String problem, the goal is to find the solution string $x^$ anywhere in $\Sigma^d$. For binary strings, the exhaustive search algorithm runs in time $O(2^d poly(nd))$ and we prove that it cannot be improved to time $O(2^{(1-\epsilon) d} poly(nd))$, for any $\epsilon > 0$, unless the Strong Exponential Time Hypothesis fails. $\bullet$ In the discrete Closest String problem, $x^$ is required to be in the input set $X$. While this problem is clearly in polynomial time, its fine-grained complexity has been pinpointed to be quadratic time $n^{2 \pm o(1)}$ whenever the dimension is $\omega(\log n) < d < n^{o(1)}$. We complement this known hardness result with new algorithms, proving essentially that whenever $d$ falls out of this hard range, the discrete Closest String problem can be solved faster than exhaustive search. In the small-$d$ regime, our algorithm is based on a novel application of the inclusion-exclusion principle. Interestingly, all of our results apply (and some are even stronger) to the natural dual of the Closest String problem, called the \emph{Remotest String} problem, where the task is to find a string maximizing the Hamming distance to all the strings in $X$.
Local Search, Semantics, and Genetic Programming: a Global Analysis
Authors: Fabio Anselmi, Mauro Castelli, Alberto d'Onofrio, Luca Manzoni, Luca Mariot, Martina Saletta
Subjects: Neural and Evolutionary Computing (cs.NE)
Abstract
Geometric Semantic Geometric Programming (GSGP) is one of the most prominent Genetic Programming (GP) variants, thanks to its solid theoretical background, the excellent performance achieved, and the execution time significantly smaller than standard syntax-based GP. In recent years, a new mutation operator, Geometric Semantic Mutation with Local Search (GSM-LS), has been proposed to include a local search step in the mutation process based on the idea that performing a linear regression during the mutation can allow for a faster convergence to good-quality solutions. While GSM-LS helps the convergence of the evolutionary search, it is prone to overfitting. Thus, it was suggested to use GSM-LS only for a limited number of generations and, subsequently, to switch back to standard geometric semantic mutation. A more recently defined variant of GSGP (called GSGP-reg) also includes a local search step but shares similar strengths and weaknesses with GSM-LS. Here we explore multiple possibilities to limit the overfitting of GSM-LS and GSGP-reg, ranging from adaptive methods to estimate the risk of overfitting at each mutation to a simple regularized regression. The results show that the method used to limit overfitting is not that important: providing that a technique to control overfitting is used, it is possible to consistently outperform standard GSGP on both training and unseen data. The obtained results allow practitioners to better understand the role of local search in GSGP and demonstrate that simple regularization strategies are effective in controlling overfitting.
Reinforcement Learning with Simple Sequence Priors
Authors: Tankred Saanum, Noémi Éltető, Peter Dayan, Marcel Binz, Eric Schulz
Abstract
Everything else being equal, simpler models should be preferred over more complex ones. In reinforcement learning (RL), simplicity is typically quantified on an action-by-action basis -- but this timescale ignores temporal regularities, like repetitions, often present in sequential strategies. We therefore propose an RL algorithm that learns to solve tasks with sequences of actions that are compressible. We explore two possible sources of simple action sequences: Sequences that can be learned by autoregressive models, and sequences that are compressible with off-the-shelf data compression algorithms. Distilling these preferences into sequence priors, we derive a novel information-theoretic objective that incentivizes agents to learn policies that maximize rewards while conforming to these priors. We show that the resulting RL algorithm leads to faster learning, and attains higher returns than state-of-the-art model-free approaches in a series of continuous control tasks from the DeepMind Control Suite. These priors also produce a powerful information-regularized agent that is robust to noisy observations and can perform open-loop control.
NeuManifold: Neural Watertight Manifold Reconstruction with Efficient and High-Quality Rendering Support
Authors: Xinyue Wei, Fanbo Xiang, Sai Bi, Anpei Chen, Kalyan Sunkavalli, Zexiang Xu, Hao Su
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
We present a method for generating high-quality watertight manifold meshes from multi-view input images. Existing volumetric rendering methods are robust in optimization but tend to generate noisy meshes with poor topology. Differentiable rasterization-based methods can generate high-quality meshes but are sensitive to initialization. Our method combines the benefits of both worlds; we take the geometry initialization obtained from neural volumetric fields, and further optimize the geometry as well as a compact neural texture representation with differentiable rasterizers. Through extensive experiments, we demonstrate that our method can generate accurate mesh reconstructions with faithful appearance that are comparable to previous volume rendering methods while being an order of magnitude faster in rendering. We also show that our generated mesh and neural texture reconstruction is compatible with existing graphics pipelines and enables downstream 3D applications such as simulation. Project page: https://sarahweiii.github.io/neumanifold/
Keyword: mobile
Continual Learning through Human-Robot Interaction -- Human Perceptions of a Continual Learning Robot in Repeated Interactions
Authors: Ali Ayub, Zachary De Francesco, Patrick Holthaus, Chrystopher L. Nehaniv, Kerstin Dautenhahn
Abstract
For long-term deployment in dynamic real-world environments, assistive robots must continue to learn and adapt to their environments. Researchers have developed various computational models for continual learning (CL) that can allow robots to continually learn from limited training data, and avoid forgetting previous knowledge. While these CL models can mitigate forgetting on static, systematically collected datasets, it is unclear how human users might perceive a robot that continually learns over multiple interactions with them. In this paper, we developed a system that integrates CL models for object recognition with a Fetch mobile manipulator robot and allows human participants to directly teach and test the robot over multiple sessions. We conducted an in-person study with 60 participants who interacted with our system in 300 sessions (5 sessions per participant). We conducted a between-participant study with three different CL models (3 experimental conditions) to understand human perceptions of continual learning robots over multiple sessions. Our results suggest that participants' perceptions of trust, competence, and usability of a continual learning robot significantly decrease over multiple sessions if the robot forgets previously learned objects. However, the perceived task load on participants for teaching and testing the robot remains the same over multiple sessions even if the robot forgets previously learned objects. Our results also indicate that state-of-the-art CL models might perform unreliably when applied to robots interacting with human participants. Further, continual learning robots are not perceived as very trustworthy or competent by human participants, regardless of the underlying continual learning model or the session number.
Sim-Suction: Learning a Suction Grasp Policy for Cluttered Environments Using a Synthetic Benchmark
Abstract
This paper presents Sim-Suction, a robust object-aware suction grasp policy for mobile manipulation platforms with dynamic camera viewpoints, designed to pick up unknown objects from cluttered environments. Suction grasp policies typically employ data-driven approaches, necessitating large-scale, accurately-annotated suction grasp datasets. However, the generation of suction grasp datasets in cluttered environments remains underexplored, leaving uncertainties about the relationship between the object of interest and its surroundings. To address this, we propose a benchmark synthetic dataset, Sim-Suction-Dataset, comprising 500 cluttered environments with 3.2 million annotated suction grasp poses. The efficient Sim-Suction-Dataset generation process provides novel insights by combining analytical models with dynamic physical simulations to create fast and accurate suction grasp pose annotations. We introduce Sim-Suction-Pointnet to generate robust 6D suction grasp poses by learning point-wise affordances from the Sim-Suction-Dataset, leveraging the synergy of zero-shot text-to-segmentation. Real-world experiments for picking up all objects demonstrate that Sim-Suction-Pointnet achieves success rates of 96.76%, 94.23%, and 92.39% on cluttered level 1 objects (prismatic shape), cluttered level 2 objects (more complex geometry), and cluttered mixed objects, respectively. The Sim-Suction policies outperform state-of-the-art benchmarks tested by approximately 21% in cluttered mixed scenes.
Digital Twin-Based 3D Map Management for Edge-Assisted Mobile Augmented Reality
Authors: Conghao Zhou, Jie Gao, Mushu Li, Nan Cheng, Xuemin Shen, Weihua Zhuang
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
In this paper, we design a 3D map management scheme for edge-assisted mobile augmented reality (MAR) to support the pose estimation of individual MAR device, which uploads camera frames to an edge server. Our objective is to minimize the pose estimation uncertainty of the MAR device by periodically selecting a proper set of camera frames for uploading to update the 3D map. To address the challenges of the dynamic uplink data rate and the time-varying pose of the MAR device, we propose a digital twin (DT)-based approach to 3D map management. First, a DT is created for the MAR device, which emulates 3D map management based on predicting subsequent camera frames. Second, a model-based reinforcement learning (MBRL) algorithm is developed, utilizing the data collected from both the actual and the emulated data to manage the 3D map. With extensive emulated data provided by the DT, the MBRL algorithm can quickly provide an adaptive map management policy in a highly dynamic environment. Simulation results demonstrate that the proposed DT-based 3D map management outperforms benchmark schemes by achieving lower pose estimation uncertainty and higher data efficiency in dynamic environments.
Fast IDentity Online with Anonymous Credentials (FIDO-AC)
Authors: Wei-Zhu Yeoh, Michal Kepkowski, Gunnar Heide, Dali Kaafar, Lucjan Hanzlik
Abstract
Web authentication is a critical component of today's Internet and the digital world we interact with. The FIDO2 protocol enables users to leverage common devices to easily authenticate to online services in both mobile and desktop environments following the passwordless authentication approach based on cryptography and biometric verification. However, there is little to no connection between the authentication process and users' attributes. More specifically, the FIDO protocol does not specify methods that could be used to combine trusted attributes with the FIDO authentication process generically and allows users to disclose them to the relying party arbitrarily. In essence, applications requiring attributes verification (e.g. age or expiry date of a driver's license, etc.) still rely on ad-hoc approaches, not satisfying the data minimization principle and not allowing the user to vet the disclosed data. A primary recent example is the data breach on Singtel Optus, one of the major telecommunications providers in Australia, where very personal and sensitive data (e.g. passport numbers) were leaked. This paper introduces FIDO-AC, a novel framework that combines the FIDO2 authentication process with the user's digital and non-shareable identity. We show how to instantiate this framework using off-the-shelf FIDO tokens and any electronic identity document, e.g., the ICAO biometric passport (ePassport). We demonstrate the practicality of our approach by evaluating a prototype implementation of the FIDO-AC system.
Incentive Mechanism for Uncertain Tasks under Differential Privacy
Authors: Xikun Jiang, Chenhao Ying, Lei Li, Haiqin Wu, Yuan Luo, Boris Düdder
Subjects: Computer Science and Game Theory (cs.GT); Cryptography and Security (cs.CR)
Abstract
Mobile crowd sensing (MCS) has emerged as an increasingly popular sensing paradigm due to its cost-effectiveness. This approach relies on platforms to outsource tasks to participating workers when prompted by task publishers. Although incentive mechanisms have been devised to foster widespread participation in MCS, most of them focus only on static tasks (i.e., tasks for which the timing and type are known in advance) and do not protect the privacy of worker bids. In a dynamic and resource-constrained environment, tasks are often uncertain (i.e., the platform lacks a priori knowledge about the tasks) and worker bids may be vulnerable to inference attacks. This paper presents HERALD, an incentive mechanism that addresses these issues through the use of uncertainty and hidden bids. Theoretical analysis reveals that HERALD satisfies a range of critical criteria, including truthfulness, individual rationality, differential privacy, low computational complexity, and low social cost. These properties are then corroborated through a series of evaluations.
Automation of Trimming Die Design Inspection by Zigzag Process Between AI and CAD Domains
Authors: Jinsub Lee, Tae-Hyun Kim, Sang-Hwan Jeon, Sung-Hyun Park, Sang-Hi Kim, Eun-Ho Lee, Jee-Hyong Lee
Abstract
Quality control in the manufacturing industry has improved with the use of artificial intelligence (AI). However, the manual inspection of trimming die designs, which is time-consuming and prone to errors, is still done by engineers. This study introduces an automatic design inspection system for automobile trimming dies by integrating AI modules and computer-aided design (CAD) software. The AI modules replace engineers' judgment, and the CAD software carries out operations requested by the AI modules. The inspection process involves a zigzag interaction between the AI modules and CAD software, enabling one-click operation without expert intervention. The AI modules are CAD-independent and data-efficient, making them adaptable to other CAD software. They achieve high performance even with limited training data, with an average length measurement error of only 2.4%. The inspection time is reduced to approximately one-fifth of the time required for manual inspection by experts.
Keyword: pruning
Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models
Abstract
Parameter-efficient tuning (PET) methods fit pre-trained language models (PLMs) to downstream tasks by either computing a small compressed update for a subset of model parameters, or appending and fine-tuning a small number of new model parameters to the pre-trained network. Hand-designed PET architectures from the literature perform well in practice, but have the potential to be improved via automated neural architecture search (NAS). We propose an efficient NAS method for learning PET architectures via structured and unstructured pruning. We present experiments on GLUE demonstrating the effectiveness of our algorithm and discuss how PET architectural design choices affect performance in practice.
Peeking inside Sparse Neural Networks using Multi-Partite Graph Representations
Authors: Elia Cunegatti, Doina Bucur, Giovanni Iacca
Abstract
Modern Deep Neural Networks (DNNs) have achieved very high performance at the expense of computational resources. To decrease the computational burden, several techniques have proposed to extract, from a given DNN, efficient subnetworks which are able to preserve performance while reducing the number of network parameters. The literature provides a broad set of techniques to discover such subnetworks, but few works have studied the peculiar topologies of such pruned architectures. In this paper, we propose a novel \emph{unrolled input-aware} bipartite Graph Encoding (GE) that is able to generate, for each layer in an either sparse or dense neural network, its corresponding graph representation based on its relation with the input data. We also extend it into a multipartite GE, to capture the relation between layers. Then, we leverage on topological properties to study the difference between the existing pruning algorithms and algorithm categories, as well as the relation between topologies and performance.
Improving Knowledge Distillation via Regularizing Feature Norm and Direction
Authors: Yuzhu Wang, Lechao Cheng, Manni Duan, Yongheng Wang, Zunlei Feng, Shu Kong
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Knowledge distillation (KD) exploits a large well-trained model (i.e., teacher) to train a small student model on the same dataset for the same task. Treating teacher features as knowledge, prevailing methods of knowledge distillation train student by aligning its features with the teacher's, e.g., by minimizing the KL-divergence between their logits or L2 distance between their intermediate features. While it is natural to believe that better alignment of student features to the teacher better distills teacher knowledge, simply forcing this alignment does not directly contribute to the student's performance, e.g., classification accuracy. In this work, we propose to align student features with class-mean of teacher features, where class-mean naturally serves as a strong classifier. To this end, we explore baseline techniques such as adopting the cosine distance based loss to encourage the similarity between student features and their corresponding class-means of the teacher. Moreover, we train the student to produce large-norm features, inspired by other lines of work (e.g., model pruning and domain adaptation), which find the large-norm features to be more significant. Finally, we propose a rather simple loss term (dubbed ND loss) to simultaneously (1) encourage student to produce large-\emph{norm} features, and (2) align the \emph{direction} of student features and teacher class-means. Experiments on standard benchmarks demonstrate that our explored techniques help existing KD methods achieve better performance, i.e., higher classification accuracy on ImageNet and CIFAR100 datasets, and higher detection precision on COCO dataset. Importantly, our proposed ND loss helps the most, leading to the state-of-the-art performance on these benchmarks. The source code is available at \url{https://github.com/WangYZ1608/Knowledge-Distillation-via-ND}.
Keyword: diffusion
Decomposing the Enigma: Subgoal-based Demonstration Learning for Formal Theorem Proving
Authors: Xueliang Zhao, Wenda Li, Lingpeng Kong
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Logic in Computer Science (cs.LO)
Abstract
Large language models~(LLMs) present an intriguing avenue of exploration in the domain of formal theorem proving. Nonetheless, the full utilization of these models, particularly in terms of demonstration formatting and organization, remains an underexplored area. In an endeavor to enhance the efficacy of LLMs, we introduce a subgoal-based demonstration learning framework, consisting of two primary elements: Firstly, drawing upon the insights of subgoal learning from the domains of reinforcement learning and robotics, we propose the construction of distinct subgoals for each demonstration example and refine these subgoals in accordance with the pertinent theories of subgoal learning. Secondly, we build upon recent advances in diffusion models to predict the optimal organization, simultaneously addressing two intricate issues that persist within the domain of demonstration organization: subset selection and order determination. Through the integration of subgoal-based learning methodologies, we have successfully increased the prevailing proof accuracy from 38.9\% to 44.3\% on the miniF2F benchmark. Furthermore, the adoption of diffusion models for demonstration organization can lead to an additional enhancement in accuracy to 45.5\%, or a $5\times$ improvement in sampling efficiency compared with the long-standing state-of-the-art method. Our code is available at \url{https://github.com/HKUNLP/subgoal-theorem-prover}.
DPOK: Reinforcement Learning for Fine-tuning Text-to-Image Diffusion Models
Authors: Ying Fan, Olivia Watkins, Yuqing Du, Hao Liu, Moonkyung Ryu, Craig Boutilier, Pieter Abbeel, Mohammad Ghavamzadeh, Kangwook Lee, Kimin Lee
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Learning from human feedback has been shown to improve text-to-image models. These techniques first learn a reward function that captures what humans care about in the task and then improve the models based on the learned reward function. Even though relatively simple approaches (e.g., rejection sampling based on reward scores) have been investigated, fine-tuning text-to-image models with the reward function remains challenging. In this work, we propose using online reinforcement learning (RL) to fine-tune text-to-image models. We focus on diffusion models, defining the fine-tuning task as an RL problem, and updating the pre-trained text-to-image diffusion models using policy gradient to maximize the feedback-trained reward. Our approach, coined DPOK, integrates policy optimization with KL regularization. We conduct an analysis of KL regularization for both RL fine-tuning and supervised fine-tuning. In our experiments, we show that DPOK is generally superior to supervised fine-tuning with respect to both image-text alignment and image quality.
Are Diffusion Models Vision-And-Language Reasoners?
Abstract
Text-conditioned image generation models have recently shown immense qualitative success using denoising diffusion processes. However, unlike discriminative vision-and-language models, it is a non-trivial task to subject these diffusion-based generative models to automatic fine-grained quantitative evaluation of high-level phenomena such as compositionality. Towards this goal, we perform two innovations. First, we transform diffusion-based models (in our case, Stable Diffusion) for any image-text matching (ITM) task using a novel method called DiffusionITM. Second, we introduce the Generative-Discriminative Evaluation Benchmark (GDBench) benchmark with 7 complex vision-and-language tasks, bias evaluation and detailed analysis. We find that Stable Diffusion + DiffusionITM is competitive on many tasks and outperforms CLIP on compositional tasks like like CLEVR and Winoground. We further boost its compositional performance with a transfer setup by fine-tuning on MS-COCO while retaining generative capabilities. We also measure the stereotypical bias in diffusion models, and find that Stable Diffusion 2.1 is, for the most part, less biased than Stable Diffusion 1.5. Overall, our results point in an exciting direction bringing discriminative and generative model evaluation closer. We will release code and benchmark setup soon.
ZeroAvatar: Zero-shot 3D Avatar Generation from a Single Image
Authors: Zhenzhen Weng, Zeyu Wang, Serena Yeung
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent advancements in text-to-image generation have enabled significant progress in zero-shot 3D shape generation. This is achieved by score distillation, a methodology that uses pre-trained text-to-image diffusion models to optimize the parameters of a 3D neural presentation, e.g. Neural Radiance Field (NeRF). While showing promising results, existing methods are often not able to preserve the geometry of complex shapes, such as human bodies. To address this challenge, we present ZeroAvatar, a method that introduces the explicit 3D human body prior to the optimization process. Specifically, we first estimate and refine the parameters of a parametric human body from a single image. Then during optimization, we use the posed parametric body as additional geometry constraint to regularize the diffusion model as well as the underlying density field. Lastly, we propose a UV-guided texture regularization term to further guide the completion of texture on invisible body parts. We show that ZeroAvatar significantly enhances the robustness and 3D consistency of optimization-based image-to-3D avatar generation, outperforming existing zero-shot image-to-3D methods.
Diffusion-Based Adversarial Sample Generation for Improved Stealthiness and Controllability
Authors: Haotian Xue, Alexandre Araujo, Bin Hu, Yongxin Chen
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Neural networks are known to be susceptible to adversarial samples: small variations of natural examples crafted to deliberately mislead the models. While they can be easily generated using gradient-based techniques in digital and physical scenarios, they often differ greatly from the actual data distribution of natural images, resulting in a trade-off between strength and stealthiness. In this paper, we propose a novel framework dubbed Diffusion-Based Projected Gradient Descent (Diff-PGD) for generating realistic adversarial samples. By exploiting a gradient guided by a diffusion model, Diff-PGD ensures that adversarial samples remain close to the original data distribution while maintaining their effectiveness. Moreover, our framework can be easily customized for specific tasks such as digital attacks, physical-world attacks, and style-based attacks. Compared with existing methods for generating natural-style adversarial samples, our framework enables the separation of optimizing adversarial loss from other surrogate losses (e.g., content/smoothness/style loss), making it more stable and controllable. Finally, we demonstrate that the samples generated using Diff-PGD have better transferability and anti-purification power than traditional gradient-based methods. Code will be released in https://github.com/xavihart/Diff-PGD
Seeding with Differentially Private Network Information
Authors: M. Amin Rahimian, Fang-Yi Yu, Carlos Hurtado
Subjects: Social and Information Networks (cs.SI); Computational Complexity (cs.CC); Multiagent Systems (cs.MA); Probability (math.PR); Applications (stat.AP)
Abstract
When designing interventions in public health, development, and education, decision makers rely on social network data to target a small number of people, capitalizing on peer effects and social contagion to bring about the most welfare benefits to the population. Developing new methods that are privacy-preserving for network data collection and targeted interventions is critical for designing sustainable public health and development interventions on social networks. In a similar vein, social media platforms rely on network data and information from past diffusions to organize their ad campaign and improve the efficacy of targeted advertising. Ensuring that these network operations do not violate users' privacy is critical to the sustainability of social media platforms and their ad economies. We study privacy guarantees for influence maximization algorithms when the social network is unknown, and the inputs are samples of prior influence cascades that are collected at random. Building on recent results that address seeding with costly network information, our privacy-preserving algorithms introduce randomization in the collected data or the algorithm output, and can bound each node's (or group of nodes') privacy loss in deciding whether or not their data should be included in the algorithm input. We provide theoretical guarantees of the seeding performance with a limited sample size subject to differential privacy budgets in both central and local privacy regimes. Simulations on synthetic and empirical network datasets reveal the diminishing value of network information with decreasing privacy budget in both regimes.
Confidence-Based Feature Imputation for Graphs with Partially Known Features
Authors: Daeho Um, Jiwoong Park, Seulki Park, Jin Young Choi
Subjects: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Abstract
This paper investigates a missing feature imputation problem for graph learning tasks. Several methods have previously addressed learning tasks on graphs with missing features. However, in cases of high rates of missing features, they were unable to avoid significant performance degradation. To overcome this limitation, we introduce a novel concept of channel-wise confidence in a node feature, which is assigned to each imputed channel feature of a node for reflecting certainty of the imputation. We then design pseudo-confidence using the channel-wise shortest path distance between a missing-feature node and its nearest known-feature node to replace unavailable true confidence in an actual learning process. Based on the pseudo-confidence, we propose a novel feature imputation scheme that performs channel-wise inter-node diffusion and node-wise inter-channel propagation. The scheme can endure even at an exceedingly high missing rate (e.g., 99.5\%) and it achieves state-of-the-art accuracy for both semi-supervised node classification and link prediction on various datasets containing a high rate of missing features. Codes are available at \url{https://github.com/daehoum1/pcfi}.
Higher Order Gauge Equivariant CNNs on Riemannian Manifolds and Applications
Authors: Gianfranco Cortes, Yue Yu, Robin Chen, Melissa Armstrong, David Vaillancourt, Baba C. Vemuri
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
With the advent of group equivariant convolutions in deep networks literature, spherical CNNs with $\mathsf{SO}(3)$-equivariant layers have been developed to cope with data that are samples of signals on the sphere $S^2$. One can implicitly obtain $\mathsf{SO}(3)$-equivariant convolutions on $S^2$ with significant efficiency gains by explicitly requiring gauge equivariance w.r.t. $\mathsf{SO}(2)$. In this paper, we build on this fact by introducing a higher order generalization of the gauge equivariant convolution, whose implementation is dubbed a gauge equivariant Volterra network (GEVNet). This allows us to model spatially extended nonlinear interactions within a given receptive field while still maintaining equivariance to global isometries. We prove theoretical results regarding the equivariance and construction of higher order gauge equivariant convolutions. Then, we empirically demonstrate the parameter efficiency of our model, first on computer vision benchmark data (e.g. spherical MNIST), and then in combination with a convolutional kernel network (CKN) on neuroimaging data. In the neuroimaging data experiments, the resulting two-part architecture (CKN + GEVNet) is used to automatically discriminate between patients with Lewy Body Disease (DLB), Alzheimer's Disease (AD) and Parkinson's Disease (PD) from diffusion magnetic resonance images (dMRI). The GEVNet extracts micro-architectural features within each voxel, while the CKN extracts macro-architectural features across voxels. This compound architecture is uniquely poised to exploit the intra- and inter-voxel information contained in the dMRI data, leading to improved performance over the classification results obtained from either of the individual components.
Diverse and Expressive Speech Prosody Prediction with Denoising Diffusion Probabilistic Model
Authors: Xiang Li, Songxiang Liu, Max W. Y. Lam, Zhiyong Wu, Chao Weng, Helen Meng
Subjects: Sound (cs.SD); Audio and Speech Processing (eess.AS)
Abstract
Expressive human speech generally abounds with rich and flexible speech prosody variations. The speech prosody predictors in existing expressive speech synthesis methods mostly produce deterministic predictions, which are learned by directly minimizing the norm of prosody prediction error. Its unimodal nature leads to a mismatch with ground truth distribution and harms the model's ability in making diverse predictions. Thus, we propose a novel prosody predictor based on the denoising diffusion probabilistic model to take advantage of its high-quality generative modeling and training stability. Experiment results confirm that the proposed prosody predictor outperforms the deterministic baseline on both the expressiveness and diversity of prediction results with even fewer network parameters.
Graph Neural Convection-Diffusion with Heterophily
Authors: Kai Zhao, Qiyu Kang, Yang Song, Rui She, Sijie Wang, Wee Peng Tay
Subjects: Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Abstract
Graph neural networks (GNNs) have shown promising results across various graph learning tasks, but they often assume homophily, which can result in poor performance on heterophilic graphs. The connected nodes are likely to be from different classes or have dissimilar features on heterophilic graphs. In this paper, we propose a novel GNN that incorporates the principle of heterophily by modeling the flow of information on nodes using the convection-diffusion equation (CDE). This allows the CDE to take into account both the diffusion of information due to homophily and the ``convection'' of information due to heterophily. We conduct extensive experiments, which suggest that our framework can achieve competitive performance on node classification tasks for heterophilic graphs, compared to the state-of-the-art methods. The code is available at \url{https://github.com/zknus/Graph-Diffusion-CDE}.
Negative-prompt Inversion: Fast Image Inversion for Editing with Text-guided Diffusion Models
Abstract
In image editing employing diffusion models, it is crucial to preserve the reconstruction quality of the original image while changing its style. Although existing methods ensure reconstruction quality through optimization, a drawback of these is the significant amount of time required for optimization. In this paper, we propose negative-prompt inversion, a method capable of achieving equivalent reconstruction solely through forward propagation without optimization, thereby enabling much faster editing processes. We experimentally demonstrate that the reconstruction quality of our method is comparable to that of existing methods, allowing for inversion at a resolution of 512 pixels and with 50 sampling steps within approximately 5 seconds, which is more than 30 times faster than null-text inversion. Reduction of the computation time by the proposed method further allows us to use a larger number of sampling steps in diffusion models to improve the reconstruction quality with a moderate increase in computation time.
Improved Visual Story Generation with Adaptive Context Modeling
Abstract
Diffusion models developed on top of powerful text-to-image generation models like Stable Diffusion achieve remarkable success in visual story generation. However, the best-performing approach considers historically generated results as flattened memory cells, ignoring the fact that not all preceding images contribute equally to the generation of the characters and scenes at the current stage. To address this, we present a simple method that improves the leading system with adaptive context modeling, which is not only incorporated in the encoder but also adopted as additional guidance in the sampling stage to boost the global consistency of the generated story. We evaluate our model on PororoSV and FlintstonesSV datasets and show that our approach achieves state-of-the-art FID scores on both story visualization and continuation scenarios. We conduct detailed model analysis and show that our model excels at generating semantically consistent images for stories.
CRoSS: Diffusion Model Makes Controllable, Robust and Secure Image Steganography
Abstract
Current image steganography techniques are mainly focused on cover-based methods, which commonly have the risk of leaking secret images and poor robustness against degraded container images. Inspired by recent developments in diffusion models, we discovered that two properties of diffusion models, the ability to achieve translation between two images without training, and robustness to noisy data, can be used to improve security and natural robustness in image steganography tasks. For the choice of diffusion model, we selected Stable Diffusion, a type of conditional diffusion model, and fully utilized the latest tools from open-source communities, such as LoRAs and ControlNets, to improve the controllability and diversity of container images. In summary, we propose a novel image steganography framework, named Controllable, Robust and Secure Image Steganography (CRoSS), which has significant advantages in controllability, robustness, and security compared to cover-based image steganography methods. These benefits are obtained without additional training. To our knowledge, this is the first work to introduce diffusion models to the field of image steganography. In the experimental section, we conducted detailed experiments to demonstrate the advantages of our proposed CRoSS framework in controllability, robustness, and security.
DiffusionNAG: Task-guided Neural Architecture Generation with Diffusion Models
Authors: Sohyun An, Hayeon Lee, Jaehyeong Jo, Seanie Lee, Sung Ju Hwang
Abstract
Neural Architecture Search (NAS) has emerged as a powerful technique for automating neural architecture design. However, existing NAS methods either require an excessive amount of time for repetitive training or sampling of many task-irrelevant architectures. Moreover, they lack generalization across different tasks and usually require searching for optimal architectures for each task from scratch without reusing the knowledge from the previous NAS tasks. To tackle such limitations of existing NAS methods, we propose a novel transferable task-guided Neural Architecture Generation (NAG) framework based on diffusion models, dubbed DiffusionNAG. With the guidance of a surrogate model, such as a performance predictor for a given task, our DiffusionNAG can generate task-optimal architectures for diverse tasks, including unseen tasks. DiffusionNAG is highly efficient as it generates task-optimal neural architectures by leveraging the prior knowledge obtained from the previous tasks and neural architecture distribution. Furthermore, we introduce a score network to ensure the generation of valid architectures represented as directed acyclic graphs, unlike existing graph generative models that focus on generating undirected graphs. Extensive experiments demonstrate that DiffusionNAG significantly outperforms the state-of-the-art transferable NAG model in architecture generation quality, as well as previous NAS methods on four computer vision datasets with largely reduced computational cost.
Learning to Imagine: Visually-Augmented Natural Language Generation
Authors: Tianyi Tang, Yushuo Chen, Yifan Du, Junyi Li, Wayne Xin Zhao, Ji-Rong Wen
Abstract
People often imagine relevant scenes to aid in the writing process. In this work, we aim to utilize visual information for composition in the same manner as humans. We propose a method, LIVE, that makes pre-trained language models (PLMs) Learn to Imagine for Visuallyaugmented natural language gEneration. First, we imagine the scene based on the text: we use a diffusion model to synthesize high-quality images conditioned on the input texts. Second, we use CLIP to determine whether the text can evoke the imagination in a posterior way. Finally, our imagination is dynamic, and we conduct synthesis for each sentence rather than generate only one image for an entire paragraph. Technically, we propose a novel plug-and-play fusion layer to obtain visually-augmented representations for each text. Our vision-text fusion layer is compatible with Transformerbased architecture. We have conducted extensive experiments on four generation tasks using BART and T5, and the automatic results and human evaluation demonstrate the effectiveness of our proposed method. We will release the code, model, and data at the link: https://github.com/RUCAIBox/LIVE.
Accelerating Diffusion Models for Inverse Problems through Shortcut Sampling
Abstract
Recently, diffusion models have demonstrated a remarkable ability to solve inverse problems in an unsupervised manner. Existing methods mainly focus on modifying the posterior sampling process while neglecting the potential of the forward process. In this work, we propose Shortcut Sampling for Diffusion (SSD), a novel pipeline for solving inverse problems. Instead of initiating from random noise, the key concept of SSD is to find the "Embryo", a transitional state that bridges the measurement image y and the restored image x. By utilizing the "shortcut" path of "input-Embryo-output", SSD can achieve precise and fast restoration. To obtain the Embryo in the forward process, We propose Distortion Adaptive Inversion (DA Inversion). Moreover, we apply back projection and attention injection as additional consistency constraints during the generation process. Experimentally, we demonstrate the effectiveness of SSD on several representative tasks, including super-resolution, deblurring, and colorization. Compared to state-of-the-art zero-shot methods, our method achieves competitive results with only 30 NFEs. Moreover, SSD with 100 NFEs can outperform state-of-the-art zero-shot methods in certain tasks.
ControlVideo: Adding Conditional Control for One Shot Text-to-Video Editing
Authors: Min Zhao, Rongzhen Wang, Fan Bao, Chongxuan Li, Jun Zhu
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
In this paper, we present ControlVideo, a novel method for text-driven video editing. Leveraging the capabilities of text-to-image diffusion models and ControlNet, ControlVideo aims to enhance the fidelity and temporal consistency of videos that align with a given text while preserving the structure of the source video. This is achieved by incorporating additional conditions such as edge maps, fine-tuning the key-frame and temporal attention on the source video-text pair with carefully designed strategies. An in-depth exploration of ControlVideo's design is conducted to inform future research on one-shot tuning video diffusion models. Quantitatively, ControlVideo outperforms a range of competitive baselines in terms of faithfulness and consistency while still aligning with the textual prompt. Additionally, it delivers videos with high visual realism and fidelity w.r.t. the source content, demonstrating flexibility in utilizing controls containing varying degrees of source video information, and the potential for multiple control combinations. The project page is available at \href{https://ml.cs.tsinghua.edu.cn/controlvideo/}{https://ml.cs.tsinghua.edu.cn/controlvideo/}.
Keyword: dynamic
Continual Learning through Human-Robot Interaction -- Human Perceptions of a Continual Learning Robot in Repeated Interactions
Authors: Ali Ayub, Zachary De Francesco, Patrick Holthaus, Chrystopher L. Nehaniv, Kerstin Dautenhahn
Abstract
For long-term deployment in dynamic real-world environments, assistive robots must continue to learn and adapt to their environments. Researchers have developed various computational models for continual learning (CL) that can allow robots to continually learn from limited training data, and avoid forgetting previous knowledge. While these CL models can mitigate forgetting on static, systematically collected datasets, it is unclear how human users might perceive a robot that continually learns over multiple interactions with them. In this paper, we developed a system that integrates CL models for object recognition with a Fetch mobile manipulator robot and allows human participants to directly teach and test the robot over multiple sessions. We conducted an in-person study with 60 participants who interacted with our system in 300 sessions (5 sessions per participant). We conducted a between-participant study with three different CL models (3 experimental conditions) to understand human perceptions of continual learning robots over multiple sessions. Our results suggest that participants' perceptions of trust, competence, and usability of a continual learning robot significantly decrease over multiple sessions if the robot forgets previously learned objects. However, the perceived task load on participants for teaching and testing the robot remains the same over multiple sessions even if the robot forgets previously learned objects. Our results also indicate that state-of-the-art CL models might perform unreliably when applied to robots interacting with human participants. Further, continual learning robots are not perceived as very trustworthy or competent by human participants, regardless of the underlying continual learning model or the session number.
OlaGPT: Empowering LLMs With Human-like Problem-Solving Abilities
Authors: Yuanzhen Xie, Tao Xie, Mingxiong Lin, WenTao Wei, Chenglin Li, Beibei Kong, Lei Chen, Chengxiang Zhuo, Bo Hu, Zang Li
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
In most current research, large language models (LLMs) are able to perform reasoning tasks by generating chains of thought through the guidance of specific prompts. However, there still exists a significant discrepancy between their capability in solving complex reasoning problems and that of humans. At present, most approaches focus on chains of thought (COT) and tool use, without considering the adoption and application of human cognitive frameworks. It is well-known that when confronting complex reasoning challenges, humans typically employ various cognitive abilities, and necessitate interaction with all aspects of tools, knowledge, and the external environment information to accomplish intricate tasks. This paper introduces a novel intelligent framework, referred to as OlaGPT. OlaGPT carefully studied a cognitive architecture framework, and propose to simulate certain aspects of human cognition. The framework involves approximating different cognitive modules, including attention, memory, reasoning, learning, and corresponding scheduling and decision-making mechanisms. Inspired by the active learning mechanism of human beings, it proposes a learning unit to record previous mistakes and expert opinions, and dynamically refer to them to strengthen their ability to solve similar problems. The paper also outlines common effective reasoning frameworks for human problem-solving and designs Chain-of-Thought (COT) templates accordingly. A comprehensive decision-making mechanism is also proposed to maximize model accuracy. The efficacy of OlaGPT has been stringently evaluated on multiple reasoning datasets, and the experimental outcomes reveal that OlaGPT surpasses state-of-the-art benchmarks, demonstrating its superior performance. Our implementation of OlaGPT is available on GitHub: \url{https://github.com/oladata-team/OlaGPT}.
Modeling Task Relationships in Multi-variate Soft Sensor with Balanced Mixture-of-Experts
Abstract
Accurate estimation of multiple quality variables is critical for building industrial soft sensor models, which have long been confronted with data efficiency and negative transfer issues. Methods sharing backbone parameters among tasks address the data efficiency issue; however, they still fail to mitigate the negative transfer problem. To address this issue, a balanced Mixture-of-Experts (BMoE) is proposed in this work, which consists of a multi-gate mixture of experts (MMoE) module and a task gradient balancing (TGB) module. The MoE module aims to portray task relationships, while the TGB module balances the gradients among tasks dynamically. Both of them cooperate to mitigate the negative transfer problem. Experiments on the typical sulfur recovery unit demonstrate that BMoE models task relationship and balances the training process effectively, and achieves better performance than baseline models significantly.
Sim-Suction: Learning a Suction Grasp Policy for Cluttered Environments Using a Synthetic Benchmark
Abstract
This paper presents Sim-Suction, a robust object-aware suction grasp policy for mobile manipulation platforms with dynamic camera viewpoints, designed to pick up unknown objects from cluttered environments. Suction grasp policies typically employ data-driven approaches, necessitating large-scale, accurately-annotated suction grasp datasets. However, the generation of suction grasp datasets in cluttered environments remains underexplored, leaving uncertainties about the relationship between the object of interest and its surroundings. To address this, we propose a benchmark synthetic dataset, Sim-Suction-Dataset, comprising 500 cluttered environments with 3.2 million annotated suction grasp poses. The efficient Sim-Suction-Dataset generation process provides novel insights by combining analytical models with dynamic physical simulations to create fast and accurate suction grasp pose annotations. We introduce Sim-Suction-Pointnet to generate robust 6D suction grasp poses by learning point-wise affordances from the Sim-Suction-Dataset, leveraging the synergy of zero-shot text-to-segmentation. Real-world experiments for picking up all objects demonstrate that Sim-Suction-Pointnet achieves success rates of 96.76%, 94.23%, and 92.39% on cluttered level 1 objects (prismatic shape), cluttered level 2 objects (more complex geometry), and cluttered mixed objects, respectively. The Sim-Suction policies outperform state-of-the-art benchmarks tested by approximately 21% in cluttered mixed scenes.
Scan and Snap: Understanding Training Dynamics and Token Composition in 1-layer Transformer
Authors: Yuandong Tian, Yiping Wang, Beidi Chen, Simon Du
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
Transformer architecture has shown impressive performance in multiple research domains and has become the backbone of many neural network models. However, there is limited understanding on how it works. In particular, with a simple predictive loss, how the representation emerges from the gradient \emph{training dynamics} remains a mystery. In this paper, for 1-layer transformer with one self-attention layer plus one decoder layer, we analyze its SGD training dynamics for the task of next token prediction in a mathematically rigorous manner. We open the black box of the dynamic process of how the self-attention layer combines input tokens, and reveal the nature of underlying inductive bias. More specifically, with the assumption (a) no positional encoding, (b) long input sequence, and (c) the decoder layer learns faster than the self-attention layer, we prove that self-attention acts as a \emph{discriminative scanning algorithm}: starting from uniform attention, it gradually attends more to distinct key tokens for a specific next token to be predicted, and pays less attention to common key tokens that occur across different next tokens. Among distinct tokens, it progressively drops attention weights, following the order of low to high co-occurrence between the key and the query token in the training set. Interestingly, this procedure does not lead to winner-takes-all, but decelerates due to a \emph{phase transition} that is controllable by the learning rates of the two layers, leaving (almost) fixed token combination. We verify this \textbf{\emph{scan and snap}} dynamics on synthetic and real-world data (WikiText).
Automatic Extraction of Time-windowed ROS Computation Graphs from ROS Bag Files
Authors: Zhuojun Chen, Michel Albonico, Ivano Malvolta
Abstract
Robotic systems react to different environmental stimuli, potentially resulting in the dynamic reconfiguration of the software controlling such systems. One effect of such dynamism is the reconfiguration of the software architecture reconfiguration of the system at runtime. Such reconfigurations might severely impact the runtime properties of robotic systems, e.g., in terms of performance and energy efficiency. The ROS \emph{rosbag} package enables developers to record and store timestamped data related to the execution of robotic missions, implicitly containing relevant information about the architecture of the monitored system during its execution. In this study, we discuss about our approach for statically extracting (time-windowed) architectural information from ROS bag files. The proposed approach can support the robotics community in better discussing and reasoning the software architecture (and its runtime reconfigurations) of ROS-based systems. We evaluate our approach against hundreds of ROS bag files systematically mined from 4,434 public GitHub repositories.
NODDLE: Node2vec based deep learning model for link prediction
Authors: Kazi Zainab Khanam, Aditya Singhal, Vijay Mago
Subjects: Social and Information Networks (cs.SI); Artificial Intelligence (cs.AI)
Abstract
Computing the probability of an edge's existence in a graph network is known as link prediction. While traditional methods calculate the similarity between two given nodes in a static network, recent research has focused on evaluating networks that evolve dynamically. Although deep learning techniques and network representation learning algorithms, such as node2vec, show remarkable improvements in prediction accuracy, the Stochastic Gradient Descent (SGD) method of node2vec tends to fall into a mediocre local optimum value due to a shortage of prior network information, resulting in failure to capture the global structure of the network. To tackle this problem, we propose NODDLE (integration of NOde2vec anD Deep Learning mEthod), a deep learning model which incorporates the features extracted by node2vec and feeds them into a four layer hidden neural network. NODDLE takes advantage of adaptive learning optimizers such as Adam, Adamax, Adadelta, and Adagrad to improve the performance of link prediction. Experimental results show that this method yields better results than the traditional methods on various social network datasets.
Neural (Tangent Kernel) Collapse
Authors: Mariia Seleznova, Dana Weitzner, Raja Giryes, Gitta Kutyniok, Hung-Hsu Chou
Abstract
This work bridges two important concepts: the Neural Tangent Kernel (NTK), which captures the evolution of deep neural networks (DNNs) during training, and the Neural Collapse (NC) phenomenon, which refers to the emergence of symmetry and structure in the last-layer features of well-trained classification DNNs. We adopt the natural assumption that the empirical NTK develops a block structure aligned with the class labels, i.e., samples within the same class have stronger correlations than samples from different classes. Under this assumption, we derive the dynamics of DNNs trained with mean squared (MSE) loss and break them into interpretable phases. Moreover, we identify an invariant that captures the essence of the dynamics, and use it to prove the emergence of NC in DNNs with block-structured NTK. We provide large-scale numerical experiments on three common DNN architectures and three benchmark datasets to support our theory.
Time to Bribe: Measuring Block Construction Market
Authors: Anton Wahrstätter, Liyi Zhou, Kaihua Qin, Davor Svetinovic, Arthur Gervais
Subjects: Networking and Internet Architecture (cs.NI)
Abstract
With the emergence of Miner Extractable Value (MEV), block construction markets on blockchains have evolved into a competitive arena. Following Ethereum's transition from Proof of Work (PoW) to Proof of Stake (PoS), the Proposer Builder Separation (PBS) mechanism has emerged as the dominant force in the Ethereum block construction market. This paper presents an in-depth longitudinal study of the Ethereum block construction market, spanning from the introduction of PoS and PBS in September 2022 to May 2023. We analyze the market shares of builders and relays, their temporal changes, and the financial dynamics within the PBS system, including payments among builders and block proposers -- commonly referred to as bribes. We introduce an MEV-time law quantifying the expected MEV revenue wrt. the time elapsed since the last proposed block. We provide empirical evidence that moments of crisis (e.g. the FTX collapse, USDC stablecoin de-peg) coincide with significant spikes in MEV payments compared to the baseline. Despite the intention of the PBS architecture to enhance decentralization by separating actor roles, it remains unclear whether its design is optimal. Implicit trust assumptions and conflicts of interest may benefit particular parties and foster the need for vertical integration. MEV-Boost was explicitly designed to foster decentralization, causing the side effect of enabling risk-free sandwich extraction from unsuspecting users, potentially raising concerns for regulators.
Abstract
We present EgoHumans, a new multi-view multi-human video benchmark to advance the state-of-the-art of egocentric human 3D pose estimation and tracking. Existing egocentric benchmarks either capture single subject or indoor-only scenarios, which limit the generalization of computer vision algorithms for real-world applications. We propose a novel 3D capture setup to construct a comprehensive egocentric multi-human benchmark in the wild with annotations to support diverse tasks such as human detection, tracking, 2D/3D pose estimation, and mesh recovery. We leverage consumer-grade wearable camera-equipped glasses for the egocentric view, which enables us to capture dynamic activities like playing soccer, fencing, volleyball, etc. Furthermore, our multi-view setup generates accurate 3D ground truth even under severe or complete occlusion. The dataset consists of more than 125k egocentric images, spanning diverse scenes with a particular focus on challenging and unchoreographed multi-human activities and fast-moving egocentric views. We rigorously evaluate existing state-of-the-art methods and highlight their limitations in the egocentric scenario, specifically on multi-human tracking. To address such limitations, we propose EgoFormer, a novel approach with a multi-stream transformer architecture and explicit 3D spatial reasoning to estimate and track the human pose. EgoFormer significantly outperforms prior art by 13.6% IDF1 and 9.3 HOTA on the EgoHumans dataset.
RoLA: A Real-Time Online Lightweight Anomaly Detection System for Multivariate Time Series
Abstract
A multivariate time series refers to observations of two or more variables taken from a device or a system simultaneously over time. There is an increasing need to monitor multivariate time series and detect anomalies in real time to ensure proper system operation and good service quality. It is also highly desirable to have a lightweight anomaly detection system that considers correlations between different variables, adapts to changes in the pattern of the multivariate time series, offers immediate responses, and provides supportive information regarding detection results based on unsupervised learning and online model training. In the past decade, many multivariate time series anomaly detection approaches have been introduced. However, they are unable to offer all the above-mentioned features. In this paper, we propose RoLA, a real-time online lightweight anomaly detection system for multivariate time series based on a divide-and-conquer strategy, parallel processing, and the majority rule. RoLA employs multiple lightweight anomaly detectors to monitor multivariate time series in parallel, determine the correlations between variables dynamically on the fly, and then jointly detect anomalies based on the majority rule in real time. To demonstrate the performance of RoLA, we conducted an experiment based on a public dataset provided by the FerryBox of the One Ocean Expedition. The results show that RoLA provides satisfactory detection accuracy and lightweight performance.
Comparing Long Short-Term Memory (LSTM) and Bidirectional LSTM Deep Neural Networks for power consumption prediction
Authors: Davi Guimarães da Silva, Anderson Alvarenga de Moura Meneses
Abstract
Electric consumption prediction methods are investigated for many reasons such as decision-making related to energy efficiency as well as for anticipating demand in the energy market dynamics. The objective of the present work is the comparison between two Deep Learning models, namely the Long Short-Term Memory (LSTM) and Bi-directional LSTM (BLSTM) for univariate electric consumption Time Series (TS) short-term forecast. The Data Sets (DSs) were selected for their different contexts and scales, aiming the assessment of the models' robustness. Four DSs were used, related to the power consumption of: (a) a household in France; (b) a university building in Santar\'em, Brazil; (c) the T\'etouan city zones, in Morocco; and (c) the Singapore aggregated electric demand. The metrics RMSE, MAE, MAPE and R2 were calculated in a TS cross-validation scheme. The Friedman's test was applied to normalized RMSE (NRMSE) results, showing that BLSTM outperforms LSTM with statistically significant difference (p = 0.0455), corroborating the fact that bidirectional weight updating improves significantly the LSTM performance concerning different scales of electric power consumption.
Digital Twin-Based 3D Map Management for Edge-Assisted Mobile Augmented Reality
Authors: Conghao Zhou, Jie Gao, Mushu Li, Nan Cheng, Xuemin Shen, Weihua Zhuang
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
In this paper, we design a 3D map management scheme for edge-assisted mobile augmented reality (MAR) to support the pose estimation of individual MAR device, which uploads camera frames to an edge server. Our objective is to minimize the pose estimation uncertainty of the MAR device by periodically selecting a proper set of camera frames for uploading to update the 3D map. To address the challenges of the dynamic uplink data rate and the time-varying pose of the MAR device, we propose a digital twin (DT)-based approach to 3D map management. First, a DT is created for the MAR device, which emulates 3D map management based on predicting subsequent camera frames. Second, a model-based reinforcement learning (MBRL) algorithm is developed, utilizing the data collected from both the actual and the emulated data to manage the 3D map. With extensive emulated data provided by the DT, the MBRL algorithm can quickly provide an adaptive map management policy in a highly dynamic environment. Simulation results demonstrate that the proposed DT-based 3D map management outperforms benchmark schemes by achieving lower pose estimation uncertainty and higher data efficiency in dynamic environments.
A Multi-Resolution Physics-Informed Recurrent Neural Network: Formulation and Application to Musculoskeletal Systems
Authors: Karan Taneja, Xiaolong He, Qizhi He, J. S. Chen
Abstract
This work presents a multi-resolution physics-informed recurrent neural network (MR PI-RNN), for simultaneous prediction of musculoskeletal (MSK) motion and parameter identification of the MSK systems. The MSK application was selected as the model problem due to its challenging nature in mapping the high-frequency surface electromyography (sEMG) signals to the low-frequency body joint motion controlled by the MSK and muscle contraction dynamics. The proposed method utilizes the fast wavelet transform to decompose the mixed frequency input sEMG and output joint motion signals into nested multi-resolution signals. The prediction model is subsequently trained on coarser-scale input-output signals using a gated recurrent unit (GRU), and then the trained parameters are transferred to the next level of training with finer-scale signals. These training processes are repeated recursively under a transfer-learning fashion until the full-scale training (i.e., with unfiltered signals) is achieved, while satisfying the underlying dynamic equilibrium. Numerical examples on recorded subject data demonstrate the effectiveness of the proposed framework in generating a physics-informed forward-dynamics surrogate, which yields higher accuracy in motion predictions of elbow flexion-extension of an MSK system compared to the case with single-scale training. The framework is also capable of identifying muscle parameters that are physiologically consistent with the subject's kinematics data.
A Hybrid Neural Coding Approach for Pattern Recognition with Spiking Neural Networks
Authors: Xinyi Chen, Qu Yang, Jibin Wu, Haizhou Li, Kay Chen Tan
Subjects: Neural and Evolutionary Computing (cs.NE)
Abstract
The biological neural systems evolved to adapt to ecological environment for efficiency and effectiveness, wherein neurons with heterogeneous structures and rich dynamics are optimized to accomplish complex cognitive tasks. Most of the current research of biologically inspired spiking neural networks (SNNs) are, however, grounded on a homogeneous neural coding scheme, which limits their overall performance in terms of accuracy, latency, efficiency, and robustness, etc. In this work, we argue that one should holistically design the network architecture to incorporate diverse neuronal functions and neural coding schemes for best performance. As an early attempt in this research direction, we put forward a hybrid neural coding framework that integrates multiple neural coding schemes discovered in neuroscience. We demonstrate that the proposed hybrid coding scheme achieves a comparable accuracy with the state-of-the-art SNNs with homogeneous neural coding on CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets with less than eight time steps and at least 3.90x fewer computations. Furthermore, we demonstrate accurate, rapid, and robust sound source localization on SoClas dataset. This study yields valuable insights into the performance of various hybrid neural coding designs and hold significant implications for designing high performance SNNs.
Behavioral Patterns in a Disease Spreading Simulation
Authors: Ollin D. Langle-Chimal, Scott C. Merril, Eric M. Clark, Gabriela Bucini, Tung-Lin Liu, Trisha R. Shrum, Christopher Koliba, Asim Zia, Julia M. Smith an Nicholas Cheney
Abstract
Human behavior is a dynamic process that evolves with experience. Understanding the evolution of individual's risk propensity is critical to design public health interventions to propitiate the adoption of better biosecurity protocols and thus, prevent the transmission of an infectious disease. Using an experimental game that simulates the spread of a disease in a network of porcine farms, we measure how learning from experience affects the risk aversion of over $1000$ players. We used a fully automated approach to segment the players into 4 categories based on the temporal trends of their game plays and compare the outcomes of their overall game performance. We found that the risk tolerant group is $50\%$ more likely to incur an infection than the risk averse one. We also find that while all individuals decrease the amount of time it takes to make decisions as they become more experienced at the game, we find a group of players with constant decision strategies who rapidly decrease their time to make a decision and a second context-aware decision group that contemplates longer before decisions while presumably performing a real-time risk assessment. The behavioral strategies employed by players in this simulated setting could be used in the future as an early warning signal to identify undesirable biosecurity-related risk aversion preferences, or changes in behavior, which may allow for targeted interventions to help mitigate them.
Mitigating Exploitation Bias in Learning to Rank with an Uncertainty-aware Empirical Bayes Approach
Authors: Tao Yang, Cuize Han, Chen Luo, Parth Gupta, Jeff M. Phillips, Qingyao Ai
Abstract
Ranking is at the core of many artificial intelligence (AI) applications, including search engines, recommender systems, etc. Modern ranking systems are often constructed with learning-to-rank (LTR) models built from user behavior signals. While previous studies have demonstrated the effectiveness of using user behavior signals (e.g., clicks) as both features and labels of LTR algorithms, we argue that existing LTR algorithms that indiscriminately treat behavior and non-behavior signals in input features could lead to suboptimal performance in practice. Particularly because user behavior signals often have strong correlations with the ranking objective and can only be collected on items that have already been shown to users, directly using behavior signals in LTR could create an exploitation bias that hurts the system performance in the long run. To address the exploitation bias, we propose EBRank, an empirical Bayes-based uncertainty-aware ranking algorithm. Specifically, to overcome exploitation bias brought by behavior features in ranking models, EBRank uses a sole non-behavior feature based prior model to get a prior estimation of relevance. In the dynamic training and serving of ranking systems, EBRank uses the observed user behaviors to update posterior relevance estimation instead of concatenating behaviors as features in ranking models. Besides, EBRank additionally applies an uncertainty-aware exploration strategy to explore actively, collect user behaviors for empirical Bayesian modeling and improve ranking performance. Experiments on three public datasets show that EBRank is effective, practical and significantly outperforms state-of-the-art ranking algorithms.
Pedestrian Trajectory Forecasting Using Deep Ensembles Under Sensing Uncertainty
Abstract
One of the fundamental challenges in the prediction of dynamic agents is robustness. Usually, most predictions are deterministic estimates of future states which are over-confident and prone to error. Recently, few works have addressed capturing uncertainty during forecasting of future states. However, these probabilistic estimation methods fail to account for the upstream noise in perception data during tracking. Sensors always have noise and state estimation becomes even more difficult under adverse weather conditions and occlusion. Traditionally, Bayes filters have been used to fuse information from noisy sensors to update states with associated belief. But, they fail to address non-linearities and long-term predictions. Therefore, we propose an end-to-end estimator that can take noisy sensor measurements and make robust future state predictions with uncertainty bounds while simultaneously taking into consideration the upstream perceptual uncertainty. For the current research, we consider an encoder-decoder based deep ensemble network for capturing both perception and predictive uncertainty simultaneously. We compared the current model to other approximate Bayesian inference methods. Overall, deep ensembles provided more robust predictions and the consideration of upstream uncertainty further increased the estimation accuracy for the model.
Data-Driven Optimization for Deposition with Degradable Tools
Authors: Tony Zheng, Monimoy Bujarbaruah, Francesco Borrelli
Abstract
We present a data-driven optimization approach for robotic controlled deposition with a degradable tool. Existing methods make the assumption that the tool tip is not changing or is replaced frequently. Errors can accumulate over time as the tool wears away and this leads to poor performance in the case where the tool degradation is unaccounted for during deposition. In the proposed approach, we utilize visual and force feedback to update the unknown model parameters of our tool-tip. Subsequently, we solve a constrained finite time optimal control problem for tracking a reference deposition profile, where our robot plans with the learned tool degradation dynamics. We focus on a robotic drawing problem as an illustrative example. Using real-world experiments, we show that the error in target vs actual deposition decreases when learned degradation models are used in the control design.
Attacks on Continuous Chaos Communication and Remedies for Resource Limited Devices
Authors: Rahul Vishwakarma, Ravi Monani, Amin Rezaei, Hossein Sayadi, Mehrdad Aliasgari, Ava Hedayatipour
Abstract
The Global Wearable market is anticipated to rise at a considerable rate in the next coming years and communication is a fundamental block in any wearable device. In communication, encryption methods are being used with the aid of microcontrollers or software implementations, which are power-consuming and incorporate complex hardware implementation. Internet of Things (IoT) devices are considered as resource-constrained devices that are expected to operate with low computational power and resource utilization criteria. At the same time, recent research has shown that IoT devices are highly vulnerable to emerging security threats, which elevates the need for low-power and small-size hardware-based security countermeasures. Chaotic encryption is a method of data encryption that utilizes chaotic systems and non-linear dynamics to generate secure encryption keys. It aims to provide high-level security by creating encryption keys that are sensitive to initial conditions and difficult to predict, making it challenging for unauthorized parties to intercept and decode encrypted data. Since the discovery of chaotic equations, there have been various encryption applications associated with them. In this paper, we comprehensively analyze the physical and encryption attacks on continuous chaotic systems in resource-constrained devices and their potential remedies. To this aim, we introduce different categories of attacks of chaotic encryption. Our experiments focus on chaotic equations implemented using Chua's equation and leverages circuit architectures and provide simulations proof of remedies for different attacks. These remedies are provided to block the attackers from stealing users' information (e.g., a pulse message) with negligible cost to the power and area of the design.
The Search for Stability: Learning Dynamics of Strategic Publishers with Initial Documents
Authors: Omer Madmon, Moshe Tennenholtz
Subjects: Computer Science and Game Theory (cs.GT); Information Retrieval (cs.IR)
Abstract
We study a game-theoretic model of information retrieval, in which strategic publishers aim to maximize their chances of being ranked first by the search engine, while maintaining the integrity of their original documents. We show that the commonly used PRP ranking scheme results in an unstable environment where games often fail to reach pure Nash equilibrium. We propose the Relative Ranking Principle (RRP) as an alternative ranking principle, and introduce two ranking functions that are instances of the RRP. We provide both theoretical and empirical evidence that these methods lead to a stable search ecosystem, by providing positive results on the learning dynamics convergence. We also define the publishers' and users' welfare, and demonstrate a possible publisher-user trade-off, which highlights the complexity of determining which ranking function should be selected by the search engine designer.
DKAF: KB Arbitration for Learning Task-Oriented Dialog Systems with Dialog-KB Inconsistencies
Abstract
Task-oriented dialog (TOD) agents often ground their responses on external knowledge bases (KBs). These KBs can be dynamic and may be updated frequently. Existing approaches for learning TOD agents assume the KB snapshot contemporary to each individual dialog is available during training. However, in real-world scenarios, only the latest KB snapshot is available during training and as a result, the train dialogs may contain facts conflicting with the latest KB. These dialog-KB inconsistencies in the training data may potentially confuse the TOD agent learning algorithm. In this work, we define the novel problem of learning a TOD agent with dialog-KB inconsistencies in the training data. We propose a Dialog-KB Arbitration Framework (DKAF) which reduces the dialog-KB inconsistencies by predicting the contemporary KB snapshot for each train dialog. These predicted KB snapshots are then used for training downstream TOD agents. As there are no existing datasets with dialog-KB inconsistencies, we systematically introduce inconsistencies in two publicly available dialog datasets. We show that TOD agents trained with DKAF perform better than existing baselines on both these datasets
Localization under consistent assumptions over dynamics
Authors: Matti Pekkanen, Francesco Verdoja, Ville Kyrki
Abstract
Accurate maps are a prerequisite for virtually all autonomous vehicle tasks. Most state-of-the-art maps assume a static world, and therefore dynamic objects are filtered out of the measurements. However, this division ignores movable but non-moving, i.e. semi-static, objects, which are usually recorded in the map and treated as static objects, violating the static world assumption, causing error in the localization. In this paper, we present a method for modeling moving and movable objects for matching the map and the measurements consistently. This reduces the error resulting from inconsistent categorization and treatment of non-static measurements. A semantic segmentation network is used to categorize the measurements into static and semi-static classes, and a background subtraction-based filtering method is used to remove dynamic measurements. Experimental comparison against a state-of-the-art baseline solution using real-world data from Oxford Radar RobotCar data set shows that consistent assumptions over dynamics increase localization accuracy.
A Hierarchical Approach to Population Training for Human-AI Collaboration
Authors: Yi Loo, Chen Gong, Malika Meghjani
Subjects: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
Abstract
A major challenge for deep reinforcement learning (DRL) agents is to collaborate with novel partners that were not encountered by them during the training phase. This is specifically worsened by an increased variance in action responses when the DRL agents collaborate with human partners due to the lack of consistency in human behaviors. Recent work have shown that training a single agent as the best response to a diverse population of training partners significantly increases an agent's robustness to novel partners. We further enhance the population-based training approach by introducing a Hierarchical Reinforcement Learning (HRL) based method for Human-AI Collaboration. Our agent is able to learn multiple best-response policies as its low-level policy while at the same time, it learns a high-level policy that acts as a manager which allows the agent to dynamically switch between the low-level best-response policies based on its current partner. We demonstrate that our method is able to dynamically adapt to novel partners of different play styles and skill levels in the 2-player collaborative Overcooked game environment. We also conducted a human study in the same environment to test the effectiveness of our method when partnering with real human subjects.
Incentive Mechanism for Uncertain Tasks under Differential Privacy
Authors: Xikun Jiang, Chenhao Ying, Lei Li, Haiqin Wu, Yuan Luo, Boris Düdder
Subjects: Computer Science and Game Theory (cs.GT); Cryptography and Security (cs.CR)
Abstract
Mobile crowd sensing (MCS) has emerged as an increasingly popular sensing paradigm due to its cost-effectiveness. This approach relies on platforms to outsource tasks to participating workers when prompted by task publishers. Although incentive mechanisms have been devised to foster widespread participation in MCS, most of them focus only on static tasks (i.e., tasks for which the timing and type are known in advance) and do not protect the privacy of worker bids. In a dynamic and resource-constrained environment, tasks are often uncertain (i.e., the platform lacks a priori knowledge about the tasks) and worker bids may be vulnerable to inference attacks. This paper presents HERALD, an incentive mechanism that addresses these issues through the use of uncertainty and hidden bids. Theoretical analysis reveals that HERALD satisfies a range of critical criteria, including truthfulness, individual rationality, differential privacy, low computational complexity, and low social cost. These properties are then corroborated through a series of evaluations.
Joint Optimization of Triangle Mesh, Material, and Light from Neural Fields with Neural Radiance Cache
Abstract
Traditional inverse rendering techniques are based on textured meshes, which naturally adapts to modern graphics pipelines, but costly differentiable multi-bounce Monte Carlo (MC) ray tracing poses challenges for modeling global illumination. Recently, neural fields has demonstrated impressive reconstruction quality but falls short in modeling indirect illumination. In this paper, we introduce a simple yet efficient inverse rendering framework that combines the strengths of both methods. Specifically, given pre-trained neural field representing the scene, we can obtain an initial estimate of the signed distance field (SDF) and create a Neural Radiance Cache (NRC), an enhancement over the traditional radiance cache used in real-time rendering. By using the former to initialize differentiable marching tetrahedrons (DMTet) and the latter to model indirect illumination, we can compute the global illumination via single-bounce differentiable MC ray tracing and jointly optimize the geometry, material, and light through back propagation. Experiments demonstrate that, compared to previous methods, our approach effectively prevents indirect illumination effects from being baked into materials, thus obtaining the high-quality reconstruction of triangle mesh, Physically-Based (PBR) materials, and High Dynamic Range (HDR) light probe.
Location-aware Verification for Autonomous Truck Platooning Based on Blockchain and Zero-knowledge Proof
Abstract
Platooning technologies enable trucks to drive cooperatively and automatically, which bring benefits including less fuel consumption, more road capacity and safety. In order to establish trust during dynamic platoon formation, ensure vehicular data integrity, and guard platoons against potential attackers, it is pivotal to verify any given vehicle's identity information before granting it access to join a platoon. To address this concern in dynamic truck platooning, we present a novel location-aware and privacy-preserving verification protocol based on zero-knowledge proof and permissioned blockchain. By performing the verification process within the spatially-local area defined by a given platoon, our system can provide lower latency and communication overhead compared to a location-agnostic blockchain system. We prototype the proposed system and perform benchmark tests on the Hyperledger platform. The experimental results show that our system is suitable for real-world truck platooning.
Learning to Imagine: Visually-Augmented Natural Language Generation
Authors: Tianyi Tang, Yushuo Chen, Yifan Du, Junyi Li, Wayne Xin Zhao, Ji-Rong Wen
Abstract
People often imagine relevant scenes to aid in the writing process. In this work, we aim to utilize visual information for composition in the same manner as humans. We propose a method, LIVE, that makes pre-trained language models (PLMs) Learn to Imagine for Visuallyaugmented natural language gEneration. First, we imagine the scene based on the text: we use a diffusion model to synthesize high-quality images conditioned on the input texts. Second, we use CLIP to determine whether the text can evoke the imagination in a posterior way. Finally, our imagination is dynamic, and we conduct synthesis for each sentence rather than generate only one image for an entire paragraph. Technically, we propose a novel plug-and-play fusion layer to obtain visually-augmented representations for each text. Our vision-text fusion layer is compatible with Transformerbased architecture. We have conducted extensive experiments on four generation tasks using BART and T5, and the automatic results and human evaluation demonstrate the effectiveness of our proposed method. We will release the code, model, and data at the link: https://github.com/RUCAIBox/LIVE.
Inverse Dynamics Pretraining Learns Good Representations for Multitask Imitation
Authors: David Brandfonbrener, Ofir Nachum, Joan Bruna
Abstract
In recent years, domains such as natural language processing and image recognition have popularized the paradigm of using large datasets to pretrain representations that can be effectively transferred to downstream tasks. In this work we evaluate how such a paradigm should be done in imitation learning, where both pretraining and finetuning data are trajectories collected by experts interacting with an unknown environment. Namely, we consider a setting where the pretraining corpus consists of multitask demonstrations and the task for each demonstration is set by an unobserved latent context variable. The goal is to use the pretraining corpus to learn a low dimensional representation of the high dimensional (e.g., visual) observation space which can be transferred to a novel context for finetuning on a limited dataset of demonstrations. Among a variety of possible pretraining objectives, we argue that inverse dynamics modeling -- i.e., predicting an action given the observations appearing before and after it in the demonstration -- is well-suited to this setting. We provide empirical evidence of this claim through evaluations on a variety of simulated visuomotor manipulation problems. While previous work has attempted various theoretical explanations regarding the benefit of inverse dynamics modeling, we find that these arguments are insufficient to explain the empirical advantages often observed in our settings, and so we derive a novel analysis using a simple but general environment model.
D-CALM: A Dynamic Clustering-based Active Learning Approach for Mitigating Bias
Abstract
Despite recent advancements, NLP models continue to be vulnerable to bias. This bias often originates from the uneven distribution of real-world data and can propagate through the annotation process. Escalated integration of these models in our lives calls for methods to mitigate bias without overbearing annotation costs. While active learning (AL) has shown promise in training models with a small amount of annotated data, AL's reliance on the model's behavior for selective sampling can lead to an accumulation of unwanted bias rather than bias mitigation. However, infusing clustering with AL can overcome the bias issue of both AL and traditional annotation methods while exploiting AL's annotation efficiency. In this paper, we propose a novel adaptive clustering-based active learning algorithm, D-CALM, that dynamically adjusts clustering and annotation efforts in response to an estimated classifier error-rate. Experiments on eight datasets for a diverse set of text classification tasks, including emotion, hatespeech, dialog act, and book type detection, demonstrate that our proposed algorithm significantly outperforms baseline AL approaches with both pretrained transformers and traditional Support Vector Machines. D-CALM showcases robustness against different measures of information gain and, as evident from our analysis of label and error distribution, can significantly reduce unwanted model bias.
GRAtt-VIS: Gated Residual Attention for Auto Rectifying Video Instance Segmentation
Authors: Tanveer Hannan, Rajat Koner, Maximilian Bernhard, Suprosanna Shit, Bjoern Menze, Volker Tresp, Matthias Schubert, Thomas Seidl
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent trends in Video Instance Segmentation (VIS) have seen a growing reliance on online methods to model complex and lengthy video sequences. However, the degradation of representation and noise accumulation of the online methods, especially during occlusion and abrupt changes, pose substantial challenges. Transformer-based query propagation provides promising directions at the cost of quadratic memory attention. However, they are susceptible to the degradation of instance features due to the above-mentioned challenges and suffer from cascading effects. The detection and rectification of such errors remain largely underexplored. To this end, we introduce \textbf{GRAtt-VIS}, \textbf{G}ated \textbf{R}esidual \textbf{Att}ention for \textbf{V}ideo \textbf{I}nstance \textbf{S}egmentation. Firstly, we leverage a Gumbel-Softmax-based gate to detect possible errors in the current frame. Next, based on the gate activation, we rectify degraded features from its past representation. Such a residual configuration alleviates the need for dedicated memory and provides a continuous stream of relevant instance features. Secondly, we propose a novel inter-instance interaction using gate activation as a mask for self-attention. This masking strategy dynamically restricts the unrepresentative instance queries in the self-attention and preserves vital information for long-term tracking. We refer to this novel combination of Gated Residual Connection and Masked Self-Attention as \textbf{GRAtt} block, which can easily be integrated into the existing propagation-based framework. Further, GRAtt blocks significantly reduce the attention overhead and simplify dynamic temporal modeling. GRAtt-VIS achieves state-of-the-art performance on YouTube-VIS and the highly challenging OVIS dataset, significantly improving over previous methods. Code is available at \url{https://github.com/Tanveer81/GRAttVIS}.
PromptNER: Prompt Locating and Typing for Named Entity Recognition
Abstract
Prompt learning is a new paradigm for utilizing pre-trained language models and has achieved great success in many tasks. To adopt prompt learning in the NER task, two kinds of methods have been explored from a pair of symmetric perspectives, populating the template by enumerating spans to predict their entity types or constructing type-specific prompts to locate entities. However, these methods not only require a multi-round prompting manner with a high time overhead and computational cost, but also require elaborate prompt templates, that are difficult to apply in practical scenarios. In this paper, we unify entity locating and entity typing into prompt learning, and design a dual-slot multi-prompt template with the position slot and type slot to prompt locating and typing respectively. Multiple prompts can be input to the model simultaneously, and then the model extracts all entities by parallel predictions on the slots. To assign labels for the slots during training, we design a dynamic template filling mechanism that uses the extended bipartite graph matching between prompts and the ground-truth entities. We conduct experiments in various settings, including resource-rich flat and nested NER datasets and low-resource in-domain and cross-domain datasets. Experimental results show that the proposed model achieves a significant performance improvement, especially in the cross-domain few-shot setting, which outperforms the state-of-the-art model by +7.7% on average.
Keyword: adaptive
Robust Representation Learning with Reliable Pseudo-labels Generation via Self-Adaptive Optimal Transport for Short Text Clustering
Abstract
Short text clustering is challenging since it takes imbalanced and noisy data as inputs. Existing approaches cannot solve this problem well, since (1) they are prone to obtain degenerate solutions especially on heavy imbalanced datasets, and (2) they are vulnerable to noises. To tackle the above issues, we propose a Robust Short Text Clustering (RSTC) model to improve robustness against imbalanced and noisy data. RSTC includes two modules, i.e., pseudo-label generation module and robust representation learning module. The former generates pseudo-labels to provide supervision for the later, which contributes to more robust representations and correctly separated clusters. To provide robustness against the imbalance in data, we propose self-adaptive optimal transport in the pseudo-label generation module. To improve robustness against the noise in data, we further introduce both class-wise and instance-wise contrastive learning in the robust representation learning module. Our empirical studies on eight short text clustering datasets demonstrate that RSTC significantly outperforms the state-of-the-art models. The code is available at: https://github.com/hmllmh/RSTC.
Stecformer: Spatio-temporal Encoding Cascaded Transformer for Multivariate Long-term Time Series Forecasting
Abstract
Multivariate long-term time series forecasting is of great application across many domains, such as energy consumption and weather forecasting. With the development of transformer-based methods, the performance of multivariate long-term time series forecasting has been significantly improved, however, the study of spatial features extracting in transformer-based model is rare and the consistency of different prediction periods is unsatisfactory due to the large span. In this work, we propose a complete solution to address these problems in terms of feature extraction and target prediction. For extraction, we design an efficient spatio-temporal encoding extractor including a semi-adaptive graph to acquire sufficient spatio-temporal information. For prediction, we propose a Cascaded Decoding Predictor (CDP) to strengthen the correlation between different intervals, which can also be utilized as a generic component to improve the performance of transformer-based methods. The proposed method, termed as Spatio-temporal Encoding Cascaded Transformer (Stecformer), achieving a notable gap over the baseline model and is comparable with the state-of-the-art performance of transformer-based methods on five benchmark datasets. We hope our attempt will serve as a regular configuration in multivariate long-term time series forecasting in the future.
ADLER -- An efficient Hessian-based strategy for adaptive learning rate
Authors: Dario Balboni, Davide Bacciu
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)
Abstract
We derive a sound positive semi-definite approximation of the Hessian of deep models for which Hessian-vector products are easily computable. This enables us to provide an adaptive SGD learning rate strategy based on the minimization of the local quadratic approximation, which requires just twice the computation of a single SGD run, but performs comparably with grid search on SGD learning rates on different model architectures (CNN with and without residual connections) on classification tasks. We also compare the novel approximation with the Gauss-Newton approximation.
NODDLE: Node2vec based deep learning model for link prediction
Authors: Kazi Zainab Khanam, Aditya Singhal, Vijay Mago
Subjects: Social and Information Networks (cs.SI); Artificial Intelligence (cs.AI)
Abstract
Computing the probability of an edge's existence in a graph network is known as link prediction. While traditional methods calculate the similarity between two given nodes in a static network, recent research has focused on evaluating networks that evolve dynamically. Although deep learning techniques and network representation learning algorithms, such as node2vec, show remarkable improvements in prediction accuracy, the Stochastic Gradient Descent (SGD) method of node2vec tends to fall into a mediocre local optimum value due to a shortage of prior network information, resulting in failure to capture the global structure of the network. To tackle this problem, we propose NODDLE (integration of NOde2vec anD Deep Learning mEthod), a deep learning model which incorporates the features extracted by node2vec and feeds them into a four layer hidden neural network. NODDLE takes advantage of adaptive learning optimizers such as Adam, Adamax, Adadelta, and Adagrad to improve the performance of link prediction. Experimental results show that this method yields better results than the traditional methods on various social network datasets.
Metaheuristic planner for cooperative multi-agent wall construction with UAVs
Authors: Basel Elkhapery, Robert Pěnička, Michal Němec, Mohsin Siddiqui
Abstract
This paper introduces a wall construction planner for Unmanned Aerial Vehicles (UAVs), which uses a Greedy Randomized Adaptive Search Procedure (GRASP) metaheuristic to generate near-time-optimal building plans for even large walls within seconds. This approach addresses one of the most time-consuming and labor-intensive tasks, while also minimizing workers' safety risks. To achieve this, the wall-building problem is modeled as a variant of the Team Orienteering Problem and is formulated as Mixed-Integer Linear Programming (MILP), with added precedence and concurrence constraints that ensure bricks are built in the correct order and without collision between cooperating agents. The GRASP planner is validated in a realistic simulation and demonstrated to find solutions with similar quality as the optimal MILP, but much faster. Moreover, it outperforms all other state-of-the-art planning approaches in the majority of test cases. This paper presents a significant advancement in the field of automated wall construction, demonstrating the potential of UAVs and optimization algorithms in improving the efficiency and safety of construction projects.
Combining Gamification and Intelligent Tutoring Systems in a Serious Game for Engineering Education
Authors: Ying Tang, Ryan Hare
Subjects: Computers and Society (cs.CY); Human-Computer Interaction (cs.HC)
Abstract
We provide ongoing results from the development of a personalized learning system integrated into a serious game. Given limited instructor resources, the use of computerized systems to help tutor students offers a way to provide higher quality education and to improve educational efficacy. Personalized learning systems like the one proposed in this paper offer an accessible solution. Furthermore, by combining such a system with a serious game, students are further engaged in interacting with the system. The proposed learning system combines expert-driven structure and lesson planning with computational intelligence methods and gamification to provide students with a fun and educational experience. As the project is ongoing from past years, numerous design iterations have been made on the system based on feedback from students and classroom observations. Using computational intelligence, the system adaptively provides support to students based on data collected from both their in-game actions and by estimating their emotional state from webcam images. For our evaluation, we focus on student data gathered from in-classroom testing in relevant courses, with both educational efficacy, results and student observations. To demonstrate the effect of our proposed system, students in an early electrical engineering course were instructed to interact with the system in place of a standard lab assignment. The system would then measure and help them improve their background knowledge before allowing them to complete the lab assignment. As they played through the game, we observed their interactions with the system to gather insights for future work. Additionally, we demonstrate the system's educational efficacy through pre-post-test results from students who played the game with and without the personalized learning system.
Digital Twin-Based 3D Map Management for Edge-Assisted Mobile Augmented Reality
Authors: Conghao Zhou, Jie Gao, Mushu Li, Nan Cheng, Xuemin Shen, Weihua Zhuang
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
In this paper, we design a 3D map management scheme for edge-assisted mobile augmented reality (MAR) to support the pose estimation of individual MAR device, which uploads camera frames to an edge server. Our objective is to minimize the pose estimation uncertainty of the MAR device by periodically selecting a proper set of camera frames for uploading to update the 3D map. To address the challenges of the dynamic uplink data rate and the time-varying pose of the MAR device, we propose a digital twin (DT)-based approach to 3D map management. First, a DT is created for the MAR device, which emulates 3D map management based on predicting subsequent camera frames. Second, a model-based reinforcement learning (MBRL) algorithm is developed, utilizing the data collected from both the actual and the emulated data to manage the 3D map. With extensive emulated data provided by the DT, the MBRL algorithm can quickly provide an adaptive map management policy in a highly dynamic environment. Simulation results demonstrate that the proposed DT-based 3D map management outperforms benchmark schemes by achieving lower pose estimation uncertainty and higher data efficiency in dynamic environments.
AdaPlanner: Adaptive Planning from Feedback with Language Models
Abstract
Large language models (LLMs) have recently demonstrated the potential in acting as autonomous agents for sequential decision-making tasks. However, most existing methods either take actions greedily without planning or rely on static plans that are not adaptable to environmental feedback. Consequently, the sequential decision-making performance of LLM agents degenerates with problem complexity and plan horizons increase. We propose a closed-loop approach, AdaPlanner, which allows the LLM agent to refine its self-generated plan adaptively in response to environmental feedback. In AdaPlanner, the LLM agent adaptively refines its plan from feedback with both in-plan and out-of-plan refinement strategies. To mitigate hallucination, we develop a code-style LLM prompt structure that facilitates plan generation across a variety of tasks, environments, and agent capabilities. Furthermore, we propose a skill discovery mechanism that leverages successful plans as few-shot exemplars, enabling the agent to plan and refine with fewer task demonstrations. Our experiments in the ALFWorld and MiniWoB++ environments demonstrate that AdaPlanner outperforms state-of-the-art baselines by 3.73% and 4.11% while utilizing 2x and 600x fewer samples, respectively.
Decentralised adaptive-gain control for eliminating epidemic spreading on networks
Authors: Liam Walsh, Mengbin Ye, Brian D.O. Anderson, Zhiyong Sun
Subjects: Systems and Control (eess.SY); Dynamical Systems (math.DS)
Abstract
This paper considers the classical Susceptible--Infected--Susceptible (SIS) network epidemic model, which describes a disease spreading through $n$ nodes, with the network links governing the possible transmission pathways of the disease between nodes. We consider feedback control to eliminate the disease in scenarios where the disease would otherwise persist in an uncontrolled network. We propose a family of decentralised adaptive-gain control algorithms, in which each node has a control gain that adaptively evolves according to a differential equation, independent of the gains of other nodes. The adaptive gain is applied multiplicatively to either decrease the infection rate or increase the recovery rate. To begin, we assume all nodes are controlled, and prove that both infection rate control and recovery rate control algorithms eliminate the disease with the limiting gains being positive and finite. Then, we consider the possibility of controlling a subset of the nodes, for both the infection rate control and recovery rate control. We first identify a necessary and sufficient condition for the existence of a subset of nodes, which if controlled would result in the elimination of the disease. For a given network, there may exist several such viable subsets, and we propose an iterative algorithm to identify such a subset. Simulations are provided to demonstrate the effectiveness of the various proposed controllers.
StyleHumanCLIP: Text-guided Garment Manipulation for StyleGAN-Human
Abstract
This paper tackles text-guided control of StyleGAN for editing garments in full-body human images. Existing StyleGAN-based methods suffer from handling the rich diversity of garments and body shapes and poses. We propose a framework for text-guided full-body human image synthesis via an attention-based latent code mapper, which enables more disentangled control of StyleGAN than existing mappers. Our latent code mapper adopts an attention mechanism that adaptively manipulates individual latent codes on different StyleGAN layers under text guidance. In addition, we introduce feature-space masking at inference time to avoid unwanted changes caused by text inputs. Our quantitative and qualitative evaluations reveal that our method can control generated images more faithfully to given texts than existing methods.
Improved Visual Story Generation with Adaptive Context Modeling
Abstract
Diffusion models developed on top of powerful text-to-image generation models like Stable Diffusion achieve remarkable success in visual story generation. However, the best-performing approach considers historically generated results as flattened memory cells, ignoring the fact that not all preceding images contribute equally to the generation of the characters and scenes at the current stage. To address this, we present a simple method that improves the leading system with adaptive context modeling, which is not only incorporated in the encoder but also adopted as additional guidance in the sampling stage to boost the global consistency of the generated story. We evaluate our model on PororoSV and FlintstonesSV datasets and show that our approach achieves state-of-the-art FID scores on both story visualization and continuation scenarios. We conduct detailed model analysis and show that our model excels at generating semantically consistent images for stories.
With a Little Push, NLI Models can Robustly and Efficiently Predict Faithfulness
Authors: Julius Steen, Juri Opitz, Anette Frank, Katja Markert
Abstract
Conditional language models still generate unfaithful output that is not supported by their input. These unfaithful generations jeopardize trust in real-world applications such as summarization or human-machine interaction, motivating a need for automatic faithfulness metrics. To implement such metrics, NLI models seem attractive, since they solve a strongly related task that comes with a wealth of prior research and data. But recent research suggests that NLI models require costly additional machinery to perform reliably across datasets, e.g., by running inference on a cartesian product of input and generated sentences, or supporting them with a question-generation/answering step. In this work we show that pure NLI models can outperform more complex metrics when combining task-adaptive data augmentation with robust inference procedures. We propose: (1) Augmenting NLI training data to adapt NL inferences to the specificities of faithfulness prediction in dialogue; (2) Making use of both entailment and contradiction probabilities in NLI, and (3) Using Monte-Carlo dropout during inference. Applied to the TRUE benchmark, which combines faithfulness datasets across diverse domains and tasks, our approach strongly improves a vanilla NLI model and significantly outperforms previous work, while showing favourable computational cost.
Local Search, Semantics, and Genetic Programming: a Global Analysis
Authors: Fabio Anselmi, Mauro Castelli, Alberto d'Onofrio, Luca Manzoni, Luca Mariot, Martina Saletta
Subjects: Neural and Evolutionary Computing (cs.NE)
Abstract
Geometric Semantic Geometric Programming (GSGP) is one of the most prominent Genetic Programming (GP) variants, thanks to its solid theoretical background, the excellent performance achieved, and the execution time significantly smaller than standard syntax-based GP. In recent years, a new mutation operator, Geometric Semantic Mutation with Local Search (GSM-LS), has been proposed to include a local search step in the mutation process based on the idea that performing a linear regression during the mutation can allow for a faster convergence to good-quality solutions. While GSM-LS helps the convergence of the evolutionary search, it is prone to overfitting. Thus, it was suggested to use GSM-LS only for a limited number of generations and, subsequently, to switch back to standard geometric semantic mutation. A more recently defined variant of GSGP (called GSGP-reg) also includes a local search step but shares similar strengths and weaknesses with GSM-LS. Here we explore multiple possibilities to limit the overfitting of GSM-LS and GSGP-reg, ranging from adaptive methods to estimate the risk of overfitting at each mutation to a simple regularized regression. The results show that the method used to limit overfitting is not that important: providing that a technique to control overfitting is used, it is possible to consistently outperform standard GSGP on both training and unseen data. The obtained results allow practitioners to better understand the role of local search in GSGP and demonstrate that simple regularization strategies are effective in controlling overfitting.
Accelerating Diffusion Models for Inverse Problems through Shortcut Sampling
Abstract
Recently, diffusion models have demonstrated a remarkable ability to solve inverse problems in an unsupervised manner. Existing methods mainly focus on modifying the posterior sampling process while neglecting the potential of the forward process. In this work, we propose Shortcut Sampling for Diffusion (SSD), a novel pipeline for solving inverse problems. Instead of initiating from random noise, the key concept of SSD is to find the "Embryo", a transitional state that bridges the measurement image y and the restored image x. By utilizing the "shortcut" path of "input-Embryo-output", SSD can achieve precise and fast restoration. To obtain the Embryo in the forward process, We propose Distortion Adaptive Inversion (DA Inversion). Moreover, we apply back projection and attention injection as additional consistency constraints during the generation process. Experimentally, we demonstrate the effectiveness of SSD on several representative tasks, including super-resolution, deblurring, and colorization. Compared to state-of-the-art zero-shot methods, our method achieves competitive results with only 30 NFEs. Moreover, SSD with 100 NFEs can outperform state-of-the-art zero-shot methods in certain tasks.
Finite Time Regret Bounds for Minimum Variance Control of Autoregressive Systems with Exogenous Inputs
Authors: Rahul Singh, Akshay Mete, Avik Kar, P. R. Kumar
Subjects: Systems and Control (eess.SY); Machine Learning (cs.LG)
Abstract
Minimum variance controllers have been employed in a wide-range of industrial applications. A key challenge experienced by many adaptive controllers is their poor empirical performance in the initial stages of learning. In this paper, we address the problem of initializing them so that they provide acceptable transients, and also provide an accompanying finite-time regret analysis, for adaptive minimum variance control of an auto-regressive system with exogenous inputs (ARX). Following [3], we consider a modified version of the Certainty Equivalence (CE) adaptive controller, which we call PIECE, that utilizes probing inputs for exploration. We show that it has a $C \log T$ bound on the regret after $T$ time-steps for bounded noise, and $C\log^2 T$ in the case of sub-Gaussian noise. The simulation results demonstrate the advantage of PIECE over the algorithm proposed in [3] as well as the standard Certainty Equivalence controller especially in the initial learning phase. To the best of our knowledge, this is the first work that provides finite-time regret bounds for an adaptive minimum variance controller.
Adaptive PD Control using Deep Reinforcement Learning for Local-Remote Teleoperation with Stochastic Time Delays
Abstract
Local-remote systems allow robots to execute complex tasks in hazardous environments such as space and nuclear power stations. However, establishing accurate positional mapping between local and remote devices can be difficult due to time delays that can compromise system performance and stability. Enhancing the synchronicity and stability of local-remote systems is vital for enabling robots to interact with environments at greater distances and under highly challenging network conditions, including time delays. We introduce an adaptive control method employing reinforcement learning to tackle the time-delayed control problem. By adjusting controller parameters in real-time, this adaptive controller compensates for stochastic delays and improves synchronicity between local and remote robotic manipulators. To improve the adaptive PD controller's performance, we devise a model-based reinforcement learning approach that effectively incorporates multi-step delays into the learning framework. Utilizing this proposed technique, the local-remote system's performance is stabilized for stochastic communication time-delays of up to 290ms. Our results demonstrate that the suggested model-based reinforcement learning method surpasses the Soft-Actor Critic and augmented state Soft-Actor Critic techniques. Access the code at: https://github.com/CAV-Research-Lab/Predictive-Model-Delay-Correction
Leveraging characteristics of the output probability distribution for identifying adversarial audio examples
Authors: Matías P. Pizarro B., Dorothea Kolossa, Asja Fischer
Subjects: Sound (cs.SD); Cryptography and Security (cs.CR); Machine Learning (cs.LG); Audio and Speech Processing (eess.AS)
Abstract
Adversarial attacks represent a security threat to machine learning based automatic speech recognition (ASR) systems. To prevent such attacks we propose an adversarial example detection strategy applicable to any ASR system that predicts a probability distribution over output tokens in each time step. We measure a set of characteristics of this distribution: the median, maximum, and minimum over the output probabilities, the entropy, and the Jensen-Shannon divergence of the distributions of subsequent time steps. Then, we fit a Gaussian distribution to the characteristics observed for benign data. By computing the likelihood of incoming new audio we can distinguish malicious inputs from samples from clean data with an area under the receiving operator characteristic (AUROC) higher than 0.99, which drops to 0.98 for less-quality audio. To assess the robustness of our method we build adaptive attacks. This reduces the AUROC to 0.96 but results in more noisy adversarial clips.
D-CALM: A Dynamic Clustering-based Active Learning Approach for Mitigating Bias
Abstract
Despite recent advancements, NLP models continue to be vulnerable to bias. This bias often originates from the uneven distribution of real-world data and can propagate through the annotation process. Escalated integration of these models in our lives calls for methods to mitigate bias without overbearing annotation costs. While active learning (AL) has shown promise in training models with a small amount of annotated data, AL's reliance on the model's behavior for selective sampling can lead to an accumulation of unwanted bias rather than bias mitigation. However, infusing clustering with AL can overcome the bias issue of both AL and traditional annotation methods while exploiting AL's annotation efficiency. In this paper, we propose a novel adaptive clustering-based active learning algorithm, D-CALM, that dynamically adjusts clustering and annotation efforts in response to an estimated classifier error-rate. Experiments on eight datasets for a diverse set of text classification tasks, including emotion, hatespeech, dialog act, and book type detection, demonstrate that our proposed algorithm significantly outperforms baseline AL approaches with both pretrained transformers and traditional Support Vector Machines. D-CALM showcases robustness against different measures of information gain and, as evident from our analysis of label and error distribution, can significantly reduce unwanted model bias.
InstaGrasp: An Entirely 3D Printed Adaptive Gripper with TPU Soft Elements and Minimal Assembly Time
Abstract
Fabricating existing and popular open-source adaptive robotic grippers commonly involves using multiple professional machines, purchasing a wide range of parts, and tedious, time-consuming assembly processes. This poses a significant barrier to entry for some robotics researchers and drives others to opt for expensive commercial alternatives. To provide both parties with an easier and cheaper (under 100GBP) solution, we propose a novel adaptive gripper design where every component (with the exception of actuators and the screws that come packaged with them) can be fabricated on a hobby-grade 3D printer, via a combination of inexpensive and readily available PLA and TPU filaments. This approach means that the gripper's tendons, flexure joints and finger pads are now printed, as a replacement for traditional string-tendons and molded urethane flexures and pads. A push-fit systems results in an assembly time of under 10 minutes. The gripper design is also highly modular and requires only a few minutes to replace any part, leading to extremely user-friendly maintenance and part modifications. An extensive stress test has shown a level of durability more than suitable for research, whilst grasping experiments (with perturbations) using items from the YCB object set has also proven its mechanical adaptability to be highly satisfactory.
Keyword: quantization
KeyPosS: Plug-and-Play Facial Landmark Detection through GPS-Inspired True-Range Multilateration
Abstract
In the realm of facial analysis, accurate landmark detection is crucial for various applications, ranging from face recognition and expression analysis to animation. Conventional heatmap or coordinate regression-based techniques, however, often face challenges in terms of computational burden and quantization errors. To address these issues, we present the KeyPoint Positioning System (KeyPosS), a groundbreaking facial landmark detection framework that stands out from existing methods. For the first time, KeyPosS employs the True-range Multilateration algorithm, a technique originally used in GPS systems, to achieve rapid and precise facial landmark detection without relying on computationally intensive regression approaches. The framework utilizes a fully convolutional network to predict a distance map, which computes the distance between a Point of Interest (POI) and multiple anchor points. These anchor points are ingeniously harnessed to triangulate the POI's position through the True-range Multilateration algorithm. Notably, the plug-and-play nature of KeyPosS enables seamless integration into any decoding stage, ensuring a versatile and adaptable solution. We conducted a thorough evaluation of KeyPosS's performance by benchmarking it against state-of-the-art models on four different datasets. The results show that KeyPosS substantially outperforms leading methods in low-resolution settings while requiring a minimal time overhead. The code is available at https://github.com/zhiqic/KeyPosS.
Implementation-Efficient Finite Alphabet Decoding of Polar Codes
Authors: Philipp Mohr, Syed Aizaz Ali Shah, Gerhard Bauch
Abstract
An implementation-efficient finite alphabet decoder for polar codes relying on coarsely quantized messages and low-complexity operations is proposed. Typically, finite alphabet decoding performs concatenated compression operations on the received channel messages to aggregate compact reliability information for error correction. These compression operations or mappings can be considered as lookup tables. For polar codes, the finite alphabet decoder design boils down to constructing lookup tables for the upper and lower branches of the building blocks within the code structure. A key challenge is to realize a hardware-friendly implementation of the lookup tables. This work uses the min-sum implementation for the upper branch lookup table and, as a novelty, a computational domain implementation for the lower branch lookup table. The computational domain approach drastically reduces the number of implementation parameters. Furthermore, a restriction to uniform quantization in the lower branch allows a very hardware-friendly compression via clipping and bit-shifting. Its behavior is close to the optimal non-uniform quantization, whose implementation would require multiple high-resolution threshold comparisons. Simulation results confirm excellent performance for the developed decoder. Unlike conventional fixed-point decoders, the proposed method involves an offline design that explicitly maximizes the preserved mutual information under coarse quantization.
Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time
Abstract
Large language models(LLMs) have sparked a new wave of exciting AI applications. Hosting these models at scale requires significant memory resources. One crucial memory bottleneck for the deployment stems from the context window. It is commonly recognized that model weights are memory hungry; however, the size of key-value embedding stored during the generation process (KV cache) can easily surpass the model size. The enormous size of the KV cache puts constraints on the inference batch size, which is crucial for high throughput inference workload. Inspired by an interesting observation of the attention scores, we hypothesize the persistence of importance: only pivotal tokens, which had a substantial influence at one step, will significantly influence future generations. Based on our empirical verification and theoretical analysis around this hypothesis, we propose Scissorhands, a system that maintains the memory usage of the KV cache at a fixed budget without finetuning the model. In essence, Scissorhands manages the KV cache by storing the pivotal tokens with a higher probability. We validate that Scissorhands reduces the inference memory usage of the KV cache by up to 5X without compromising model quality. We further demonstrate that Scissorhands can be combined with 4-bit quantization, traditionally used to compress model weights, to achieve up to 20X compression.
Keyword: efficient
Think Before You Act: Decision Transformers with Internal Working Memory
Differentiable Clustering with Perturbed Spanning Forests
Stecformer: Spatio-temporal Encoding Cascaded Transformer for Multivariate Long-term Time Series Forecasting
DeepGate2: Functionality-Aware Circuit Representation Learning
Sim-Suction: Learning a Suction Grasp Policy for Cluttered Environments Using a Synthetic Benchmark
Learning Better with Less: Effective Augmentation for Sample-Efficient Visual Reinforcement Learning
SketchOGD: Memory-Efficient Continual Learning
Learning Preconditioner for Conjugate Gradient PDE Solvers
Optimized Custom Dataset for Efficient Detection of Underwater Trash
Sample Efficient Reinforcement Learning in Mixed Systems through Augmented Samples and Its Applications to Queueing Networks
AD-NEV: A Scalable Multi-level Neuroevolution Framework for Multivariate Anomaly Detection
Sliding Window Sum Algorithms for Deep Neural Networks
Efficient Computation of Quantiles over Joins
CARAMEL: A Succinct Read-Only Lookup Table via Compressed Static Functions
CVB: A Video Dataset of Cattle Visual Behaviors
ParaAMR: A Large-Scale Syntactically Diverse Paraphrase Dataset by AMR Back-Translation
Legion: Automatically Pushing the Envelope of Multi-GPU System for Billion-Scale GNN Training
Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models
Set-based Neural Network Encoding
FARA: Future-aware Ranking Algorithm for Fairness Optimization
Improving Position Encoding of Transformers for Multivariate Time Series Classification
CAILA: Concept-Aware Intra-Layer Adapters for Compositional Zero-Shot Learning
Sharpend Cosine Similarity based Neural Network for Hyperspectral Image Classification
Future-conditioned Unsupervised Pretraining for Decision Transformer
Kaczmarz-Type Method for Solving Matrix Equation $AXB=C$
PIP: Parse-Instructed Prefix for Syntactically Controlled Paraphrase Generation
Merging control in mixed traffic with safety guarantees: a safe sequencing policy with optimal motion control
Symmetric resonance based integrators and forest formulae
Parameter-Efficient Fine-Tuning without Introducing New Latency
Leveraging Domain Knowledge for Inclusive and Bias-aware Humanitarian Response Entry Classification
Towards a Common Understanding of Contributing Factors for Cross-Lingual Transfer in Multilingual Language Models: A Review
Uncertain Pose Estimation during Contact Tasks using Differentiable Contact Features
vFedSec: Efficient Secure Aggregation for Vertical Federated Learning via Secure Layer
Joint Optimization of Triangle Mesh, Material, and Light from Neural Fields with Neural Radiance Cache
Domain Aligned Prefix Averaging for Domain Generalization in Abstractive Summarization
Randomized Positional Encodings Boost Length Generalization of Transformers
Green Runner: A tool for efficient model selection from model repositories
Automation of Trimming Die Design Inspection by Zigzag Process Between AI and CAD Domains
Modelling, Analysis and Control of OmniMorph:an Omnidirectional Morphing Multi-rotor UAV
Distributional Reinforcement Learning with Dual Expectile-Quantile Regression
Peeking inside Sparse Neural Networks using Multi-Partite Graph Representations
Feature Adaptation for Sparse Linear Regression
Submodular Minimax Optimization: Finding Effective Sets
A Neural State-Space Model Approach to Efficient Speech Separation
DiffusionNAG: Task-guided Neural Architecture Generation with Diffusion Models
Meta-prediction Model for Distillation-Aware NAS on Unseen Datasets
CUQIpy -- Part I: computational uncertainty quantification for inverse problems in Python
Implementation-Efficient Finite Alphabet Decoding of Polar Codes
Training Socially Aligned Language Models in Simulated Human Society
TranSFormer: Slow-Fast Transformer for Machine Translation
slow'' branch to deal with subword sequences and a
fast'' branch to deal with longer character sequences. This model is efficient since the fast branch is very lightweight by reducing the model width, and yet provides useful fine-grained features for the slow branch. Our TranSFormer shows consistent BLEU improvements (larger than 1 BLEU point) on several machine translation benchmarks.A Tale of Two Approximations: Tightening Over-Approximation for DNN Robustness Verification via Under-Approximation
Aggregating Capacity in FL through Successive Layer Training for Computationally-Constrained Devices
Let the Flows Tell: Solving Graph Combinatorial Optimization Problems with GFlowNets
Formal Modelling for Multi-Robot Systems Under Uncertainty
Diable: Efficient Dialogue State Tracking as Operations on Tables
GLOBE-CE: A Translation-Based Approach for Global Counterfactual Explanations
Joint Antenna Selection and Beamforming for Massive MIMO-enabled Over-the-Air Federated Learning
Complete Multiparty Session Type Projection with Automata
Communication-Efficient Reinforcement Learning in Swarm Robotic Networks for Maze Exploration
Keyword: faster
WeiAvg: Federated Learning Model Aggregation Promoting Data Diversity
DeepGate2: Functionality-Aware Circuit Representation Learning
Scan and Snap: Understanding Training Dynamics and Token Composition in 1-layer Transformer
Metaheuristic planner for cooperative multi-agent wall construction with UAVs
Accelerating Value Iteration with Anchoring
A Slingshot Approach to Learning in Monotone Games
Negative-prompt Inversion: Fast Image Inversion for Editing with Text-guided Diffusion Models
Channel and Gradient-Importance Aware Device Scheduling for Over-the-Air Federated Learning
Can You Solve Closest String Faster than Exhaustive Search?
Local Search, Semantics, and Genetic Programming: a Global Analysis
Reinforcement Learning with Simple Sequence Priors
NeuManifold: Neural Watertight Manifold Reconstruction with Efficient and High-Quality Rendering Support
Keyword: mobile
Continual Learning through Human-Robot Interaction -- Human Perceptions of a Continual Learning Robot in Repeated Interactions
Sim-Suction: Learning a Suction Grasp Policy for Cluttered Environments Using a Synthetic Benchmark
Digital Twin-Based 3D Map Management for Edge-Assisted Mobile Augmented Reality
Fast IDentity Online with Anonymous Credentials (FIDO-AC)
Incentive Mechanism for Uncertain Tasks under Differential Privacy
Automation of Trimming Die Design Inspection by Zigzag Process Between AI and CAD Domains
Keyword: pruning
Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models
Peeking inside Sparse Neural Networks using Multi-Partite Graph Representations
Improving Knowledge Distillation via Regularizing Feature Norm and Direction
Keyword: diffusion
Decomposing the Enigma: Subgoal-based Demonstration Learning for Formal Theorem Proving
DPOK: Reinforcement Learning for Fine-tuning Text-to-Image Diffusion Models
Are Diffusion Models Vision-And-Language Reasoners?
ZeroAvatar: Zero-shot 3D Avatar Generation from a Single Image
Diffusion-Based Adversarial Sample Generation for Improved Stealthiness and Controllability
Seeding with Differentially Private Network Information
Confidence-Based Feature Imputation for Graphs with Partially Known Features
Higher Order Gauge Equivariant CNNs on Riemannian Manifolds and Applications
Diverse and Expressive Speech Prosody Prediction with Denoising Diffusion Probabilistic Model
Graph Neural Convection-Diffusion with Heterophily
Negative-prompt Inversion: Fast Image Inversion for Editing with Text-guided Diffusion Models
Improved Visual Story Generation with Adaptive Context Modeling
CRoSS: Diffusion Model Makes Controllable, Robust and Secure Image Steganography
DiffusionNAG: Task-guided Neural Architecture Generation with Diffusion Models
Learning to Imagine: Visually-Augmented Natural Language Generation
Accelerating Diffusion Models for Inverse Problems through Shortcut Sampling
ControlVideo: Adding Conditional Control for One Shot Text-to-Video Editing
Keyword: dynamic
Continual Learning through Human-Robot Interaction -- Human Perceptions of a Continual Learning Robot in Repeated Interactions
OlaGPT: Empowering LLMs With Human-like Problem-Solving Abilities
Modeling Task Relationships in Multi-variate Soft Sensor with Balanced Mixture-of-Experts
Sim-Suction: Learning a Suction Grasp Policy for Cluttered Environments Using a Synthetic Benchmark
Scan and Snap: Understanding Training Dynamics and Token Composition in 1-layer Transformer
Automatic Extraction of Time-windowed ROS Computation Graphs from ROS Bag Files
NODDLE: Node2vec based deep learning model for link prediction
Neural (Tangent Kernel) Collapse
Time to Bribe: Measuring Block Construction Market
EgoHumans: An Egocentric 3D Multi-Human Benchmark
RoLA: A Real-Time Online Lightweight Anomaly Detection System for Multivariate Time Series
Comparing Long Short-Term Memory (LSTM) and Bidirectional LSTM Deep Neural Networks for power consumption prediction
Digital Twin-Based 3D Map Management for Edge-Assisted Mobile Augmented Reality
A Multi-Resolution Physics-Informed Recurrent Neural Network: Formulation and Application to Musculoskeletal Systems
A Hybrid Neural Coding Approach for Pattern Recognition with Spiking Neural Networks
Behavioral Patterns in a Disease Spreading Simulation
Mitigating Exploitation Bias in Learning to Rank with an Uncertainty-aware Empirical Bayes Approach
Pedestrian Trajectory Forecasting Using Deep Ensembles Under Sensing Uncertainty
Data-Driven Optimization for Deposition with Degradable Tools
Attacks on Continuous Chaos Communication and Remedies for Resource Limited Devices
The Search for Stability: Learning Dynamics of Strategic Publishers with Initial Documents
DKAF: KB Arbitration for Learning Task-Oriented Dialog Systems with Dialog-KB Inconsistencies
Localization under consistent assumptions over dynamics
A Hierarchical Approach to Population Training for Human-AI Collaboration
Incentive Mechanism for Uncertain Tasks under Differential Privacy
Joint Optimization of Triangle Mesh, Material, and Light from Neural Fields with Neural Radiance Cache
Location-aware Verification for Autonomous Truck Platooning Based on Blockchain and Zero-knowledge Proof
Learning to Imagine: Visually-Augmented Natural Language Generation
Inverse Dynamics Pretraining Learns Good Representations for Multitask Imitation
D-CALM: A Dynamic Clustering-based Active Learning Approach for Mitigating Bias
GRAtt-VIS: Gated Residual Attention for Auto Rectifying Video Instance Segmentation
PromptNER: Prompt Locating and Typing for Named Entity Recognition
Keyword: adaptive
Robust Representation Learning with Reliable Pseudo-labels Generation via Self-Adaptive Optimal Transport for Short Text Clustering
Stecformer: Spatio-temporal Encoding Cascaded Transformer for Multivariate Long-term Time Series Forecasting
ADLER -- An efficient Hessian-based strategy for adaptive learning rate
NODDLE: Node2vec based deep learning model for link prediction
Metaheuristic planner for cooperative multi-agent wall construction with UAVs
Combining Gamification and Intelligent Tutoring Systems in a Serious Game for Engineering Education
Digital Twin-Based 3D Map Management for Edge-Assisted Mobile Augmented Reality
AdaPlanner: Adaptive Planning from Feedback with Language Models
Decentralised adaptive-gain control for eliminating epidemic spreading on networks
StyleHumanCLIP: Text-guided Garment Manipulation for StyleGAN-Human
Improved Visual Story Generation with Adaptive Context Modeling
With a Little Push, NLI Models can Robustly and Efficiently Predict Faithfulness
Local Search, Semantics, and Genetic Programming: a Global Analysis
Accelerating Diffusion Models for Inverse Problems through Shortcut Sampling
Finite Time Regret Bounds for Minimum Variance Control of Autoregressive Systems with Exogenous Inputs
Adaptive PD Control using Deep Reinforcement Learning for Local-Remote Teleoperation with Stochastic Time Delays
Leveraging characteristics of the output probability distribution for identifying adversarial audio examples
D-CALM: A Dynamic Clustering-based Active Learning Approach for Mitigating Bias
InstaGrasp: An Entirely 3D Printed Adaptive Gripper with TPU Soft Elements and Minimal Assembly Time
Keyword: quantization
KeyPosS: Plug-and-Play Facial Landmark Detection through GPS-Inspired True-Range Multilateration
Implementation-Efficient Finite Alphabet Decoding of Polar Codes
Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time