Abstract
Quantization is commonly used in Deep Neural Networks (DNNs) to reduce the storage and computational complexity by decreasing the arithmetical precision of activations and weights, a.k.a. tensors. Efficient hardware architectures employ linear quantization to enable the deployment of recent DNNs onto embedded systems and mobile devices. However, linear uniform quantization cannot usually reduce the numerical precision to less than 8 bits without sacrificing high performance in terms of model accuracy. The performance loss is due to the fact that tensors do not follow uniform distributions. In this paper, we show that a significant amount of tensors fit into an exponential distribution. Then, we propose DNA-TEQ to exponentially quantize DNN tensors with an adaptive scheme that achieves the best trade-off between numerical precision and accuracy loss. The experimental results show that DNA-TEQ provides a much lower quantization bit-width compared to previous proposals, resulting in an average compression ratio of 40% over the linear INT8 baseline, with negligible accuracy loss and without retraining the DNNs. Besides, DNA-TEQ leads the way in performing dot-product operations in the exponential domain, which saves 66% of energy consumption on average for a set of widely used DNNs.
DenseBAM-GI: Attention Augmented DeneseNet with momentum aided GRU for HMER
Authors: Aniket Pal, Krishna Pratap Singh
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
The task of recognising Handwritten Mathematical Expressions (HMER) is crucial in the fields of digital education and scholarly research. However, it is difficult to accurately determine the length and complex spatial relationships among symbols in handwritten mathematical expressions. In this study, we present a novel encoder-decoder architecture (DenseBAM-GI) for HMER, where the encoder has a Bottleneck Attention Module (BAM) to improve feature representation and the decoder has a Gated Input-GRU (GI-GRU) unit with an extra gate to make decoding long and complex expressions easier. The proposed model is an efficient and lightweight architecture with performance equivalent to state-of-the-art models in terms of Expression Recognition Rate (exprate). It also performs better in terms of top 1, 2, and 3 error accuracy across the CROHME 2014, 2016, and 2019 datasets. DenseBAM-GI achieves the best exprate among all models on the CROHME 2019 dataset. Importantly, these successes are accomplished with a drop in the complexity of the calculation and a reduction in the need for GPU memory.
Equivariant Neural Networks for Indirect Measurements
Authors: Matthias Beckmann, Nick Heilenkötter
Subjects: Numerical Analysis (math.NA); Information Theory (cs.IT); Functional Analysis (math.FA)
Abstract
In the recent years, deep learning techniques have shown great success in various tasks related to inverse problems, where a target quantity of interest can only be observed through indirect measurements by a forward operator. Common approaches apply deep neural networks in a post-processing step to the reconstructions obtained by classical reconstruction methods. However, the latter methods can be computationally expensive and introduce artifacts that are not present in the measured data and, in turn, can deteriorate the performance on the given task. To overcome these limitations, we propose a class of equivariant neural networks that can be directly applied to the measurements to solve the desired task. To this end, we build appropriate network structures by developing layers that are equivariant with respect to data transformations induced by well-known symmetries in the domain of the forward operator. We rigorously analyze the relation between the measurement operator and the resulting group representations and prove a representer theorem that characterizes the class of linear operators that translate between a given pair of group actions. Based on this theory, we extend the existing concepts of Lie group equivariant deep learning to inverse problems and introduce new representations that result from the involved measurement operations. This allows us to efficiently solve classification, regression or even reconstruction tasks based on indirect measurements also for very sparse data problems, where a classical reconstruction-based approach may be hard or even impossible. We illustrate the effectiveness of our approach in numerical experiments and compare with existing methods.
Solver algorithm for stabilized space-time formulation of advection-dominated diffusion problem
Authors: Marcin Łoś, Paulina Sepulveda-Salas, Maciej Paszyński
Abstract
This article shows how to develop an efficient solver for a stabilized numerical space-time formulation of the advection-dominated diffusion transient equation. At the discrete space-time level, we approximate the solution by using higher-order continuous B-spline basis functions in its spatial and temporal dimensions. This problem is very difficult to solve numerically using the standard Galerkin finite element method due to artificial oscillations present when the advection term dominates the diffusion term. However, a first-order constraint least-square formulation allows us to obtain numerical solutions avoiding oscillations. The advantages of space-time formulations are the use of high-order methods and the feasibility of developing space-time mesh adaptive techniques on well-defined discrete problems. We develop a solver for a least-square formulation to obtain a stabilized and symmetric problem on finite element meshes. The computational cost of our solver is bounded by the cost of the inversion of the space-time mass and stiffness (with one value fixed at a point) matrices and the cost of the GMRES solver applied for the symmetric and positive definite problem. We illustrate our findings on an advection-dominated diffusion space-time model problem and present two numerical examples: one with isogeometric analysis discretizations and the second one with an adaptive space-time finite element method.
HNO: Hyena Neural Operator for solving PDEs
Authors: Saurabh Patil, Zijie Li, Amir Barati Farimani
Abstract
Numerically solving partial differential equations (PDEs) typically requires fine discretization to resolve necessary spatiotemporal scales, which can be computationally expensive. Recent advances in deep learning have provided a new approach to solving PDEs that involves the use of neural operators. Neural operators are neural network architectures that learn mappings between function spaces and have the capability to solve partial differential equations based on data. This study utilizes a novel neural operator called Hyena, which employs a long convolutional filter that is parameterized by a multilayer perceptron. The Hyena operator is an operation that enjoys sub-quadratic complexity and state space model to parameterize long convolution that enjoys global receptive field. This mechanism enhances the model's comprehension of the input's context and enables data-dependent weight for different PDE instances. To measure how effective the layers are in solving PDEs, we conduct experiments on Burger's equation and Navier Stokes equation. Our findings indicate Hyena Neural operator can serve as an efficient and accurate model for learning PDEs' solution operator. The data and code used can be found at: https://github.com/Saupatil07/Hyena-Neural-Operator
CLANet: A Comprehensive Framework for Cross-Batch Cell Line Identification Using Brightfield Images
Authors: Lei Tong, Adam Corrigan, Navin Rathna Kumar, Kerry Hallbrook, Jonathan Orme, Yinhai Wang, Huiyu Zhou
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Cell line authentication plays a crucial role in the biomedical field, ensuring researchers work with accurately identified cells. Supervised deep learning has made remarkable strides in cell line identification by studying cell morphological features through cell imaging. However, batch effects, a significant issue stemming from the different times at which data is generated, lead to substantial shifts in the underlying data distribution, thus complicating reliable differentiation between cell lines from distinct batch cultures. To address this challenge, we introduce CLANet, a pioneering framework for cross-batch cell line identification using brightfield images, specifically designed to tackle three distinct batch effects. We propose a cell cluster-level selection method to efficiently capture cell density variations, and a self-supervised learning strategy to manage image quality variations, thus producing reliable patch representations. Additionally, we adopt multiple instance learning(MIL) for effective aggregation of instance-level features for cell line identification. Our innovative time-series segment sampling module further enhances MIL's feature-learning capabilities, mitigating biases from varying incubation times across batches. We validate CLANet using data from 32 cell lines across 93 experimental batches from the AstraZeneca Global Cell Bank. Our results show that CLANet outperforms related approaches (e.g. domain adaptation, MIL), demonstrating its effectiveness in addressing batch effects in cell line identification.
Envisioning a Next Generation Extended Reality Conferencing System with Efficient Photorealistic Human Rendering
Abstract
Meeting online is becoming the new normal. Creating an immersive experience for online meetings is a necessity towards more diverse and seamless environments. Efficient photorealistic rendering of human 3D dynamics is the core of immersive meetings. Current popular applications achieve real-time conferencing but fall short in delivering photorealistic human dynamics, either due to limited 2D space or the use of avatars that lack realistic interactions between participants. Recent advances in neural rendering, such as the Neural Radiance Field (NeRF), offer the potential for greater realism in metaverse meetings. However, the slow rendering speed of NeRF poses challenges for real-time conferencing. We envision a pipeline for a future extended reality metaverse conferencing system that leverages monocular video acquisition and free-viewpoint synthesis to enhance data and hardware efficiency. Towards an immersive conferencing experience, we explore an accelerated NeRF-based free-viewpoint synthesis algorithm for rendering photorealistic human dynamics more efficiently. We show that our algorithm achieves comparable rendering quality while performing training and inference 44.5% and 213% faster than state-of-the-art methods, respectively. Our exploration provides a design basis for constructing metaverse conferencing systems that can handle complex application scenarios, including dynamic scene relighting with customized themes and multi-user conferencing that harmonizes real-world people into an extended world.
Performance Analysis of Empirical Open-Circuit Voltage Modeling in Lithium Ion Batteries, Part-1: Performance Measures
Authors: Prarthana Pillai, James Nguyen, Balakumar Balasingam
Abstract
The open circuit voltage to the state of charge (OCVSOC) characteristic is crucial for battery management systems. Using the OCV-SOC curve, the SOC and the battery capacity can be estimated in real-time. Accurate SOC and capacity information are important to carry out the majority of battery management functionalities that ensure a safe, efficient, and reliable battery pack power system. Numerous approaches have been reported in the literature for improved SOC estimation and battery capacity estimation. These approaches focus on various estimation and filtering techniques to reduce the effect of measurement noise and uncertainties due to hysteresis and relaxation effects. Even though all the existing approaches to SOC estimation rely on the OCV-SOC characterization, little attention was paid to investigating the possibility of errors in the OCV-SOC characterization and the effect of uncertainty in the OCV-SOC curve on SOC and capacity estimates. In this paper, which is the first part of a series of three papers, the effect of OCV-SOC modeling error in the overall battery management system is discussed. The different sources of uncertainties in the OCV-SOC curve include cell-to-cell variation, temperature variation, aging drift, cycle rate effect, curve-fitting error, and measurement/estimation error. The proposed uncertainty models can be incorporated into battery management systems to improve their safety, performance, and reliability.
Performance Analysis of Empirical Open-Circuit Voltage Modeling in Lithium Ion Batteries, Part-2: Data Collection Procedure
Authors: Prarthana Pillai, James Nguyen, Balakumar Balasingam
Abstract
This paper is the second part of a series of papers about empirical approaches to open circuit voltage (OCV) modeling and its performance comparison in lithium-ion batteries. The first part of the series introduced various sources of uncertainties in the OCV models and established a theoretical relationship between uncertainties and the performance of a battery management system. In this paper, clearly defined approaches for low-rate OCV data collection are defined and described in detail. The data collection is designed with consideration to several parameters that affect the experimental time. Firstly, a more suitable method to fully charge the battery at different C-Rates is defined. Secondly, the OCV characterization following the full charge is described for various performance comparisons. Finally, optimal and efficient resistance estimation profiles are discussed. From the voltage, current and time data recorded using the procedure described in this paper, the OCV-SOC relationship is characterized and its uncertainties are modeled in the third part of this series of papers.
An Efficient Sparse Inference Software Accelerator for Transformer-based Language Models on CPUs
Authors: Haihao Shen, Hengyu Meng, Bo Dong, Zhe Wang, Ofir Zafrir, Yi Ding, Yu Luo, Hanwen Chang, Qun Gao, Ziheng Wang, Guy Boudoukh, Moshe Wasserblat
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
In recent years, Transformer-based language models have become the standard approach for natural language processing tasks. However, stringent throughput and latency requirements in industrial applications are limiting their adoption. To mitigate the gap, model compression techniques such as structured pruning are being used to improve inference efficiency. However, most existing neural network inference runtimes lack adequate support for structured sparsity. In this paper, we propose an efficient sparse deep learning inference software stack for Transformer-based language models where the weights are pruned with constant block size. Our sparse software accelerator leverages Intel Deep Learning Boost to maximize the performance of sparse matrix - dense matrix multiplication (commonly abbreviated as SpMM) on CPUs. Our SpMM kernel outperforms the existing sparse libraries (oneMKL, TVM, and LIBXSMM) by an order of magnitude on a wide range of GEMM shapes under 5 representative sparsity ratios (70%, 75%, 80%, 85%, 90%). Moreover, our SpMM kernel shows up to 5x speedup over dense GEMM kernel of oneDNN, a well-optimized dense library widely used in industry. We apply our sparse accelerator on widely-used Transformer-based language models including Bert-Mini, DistilBERT, Bert-Base, and BERT-Large. Our sparse inference software shows up to 1.5x speedup over Neural Magic's Deepsparse under same configurations on Xeon on Amazon Web Services under proxy production latency constraints. We also compare our solution with two framework-based inference solutions, ONNX Runtime and PyTorch, and demonstrate up to 37x speedup over ONNX Runtime and 345x over PyTorch on Xeon under the latency constraints. All the source code is publicly available on Github: https://github.com/intel/intel-extension-for-transformers.
GuidedMixup: An Efficient Mixup Strategy Guided by Saliency Maps
Abstract
Data augmentation is now an essential part of the image training process, as it effectively prevents overfitting and makes the model more robust against noisy datasets. Recent mixing augmentation strategies have advanced to generate the mixup mask that can enrich the saliency information, which is a supervisory signal. However, these methods incur a significant computational burden to optimize the mixup mask. From this motivation, we propose a novel saliency-aware mixup method, GuidedMixup, which aims to retain the salient regions in mixup images with low computational overhead. We develop an efficient pairing algorithm that pursues to minimize the conflict of salient regions of paired images and achieve rich saliency in mixup images. Moreover, GuidedMixup controls the mixup ratio for each pixel to better preserve the salient region by interpolating two paired images smoothly. The experiments on several datasets demonstrate that GuidedMixup provides a good trade-off between augmentation overhead and generalization performance on classification datasets. In addition, our method shows good performance in experiments with corrupted or reduced datasets.
Group-based Robustness: A General Framework for Customized Robustness in the Real World
Authors: Weiran Lin, Keane Lucas, Neo Eyal, Lujo Bauer, Michael K. Reiter, Mahmood Sharif
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Machine-learning models are known to be vulnerable to evasion attacks that perturb model inputs to induce misclassifications. In this work, we identify real-world scenarios where the true threat cannot be assessed accurately by existing attacks. Specifically, we find that conventional metrics measuring targeted and untargeted robustness do not appropriately reflect a model's ability to withstand attacks from one set of source classes to another set of target classes. To address the shortcomings of existing methods, we formally define a new metric, termed group-based robustness, that complements existing metrics and is better-suited for evaluating model performance in certain attack scenarios. We show empirically that group-based robustness allows us to distinguish between models' vulnerability against specific threat models in situations where traditional robustness metrics do not apply. Moreover, to measure group-based robustness efficiently and accurately, we 1) propose two loss functions and 2) identify three new attack strategies. We show empirically that with comparable success rates, finding evasive samples using our new loss functions saves computation by a factor as large as the number of targeted classes, and finding evasive samples using our new attack strategies saves time by up to 99\% compared to brute-force search methods. Finally, we propose a defense method that increases group-based robustness by up to 3.52$\times$.
Entanglement Routing over Quantum Networks Using Greenberger-Horne-Zeilinger Measurements
Authors: Yiming Zeng, Jiarui Zhang, Ji Liu, Zhenhua Liu, Yuanyuan Yang
Subjects: Networking and Internet Architecture (cs.NI)
Abstract
Generating a long-distance quantum entanglement is one of the most essential functions of a quantum network to support quantum communication and computing applications. The successful entanglement rate during a probabilistic entanglement process decreases dramatically with distance, and swapping is a widely-applied quantum technique to address this issue. Most existing entanglement routing protocols use a classic entanglement-swapping method based on Bell State measurements that can only fuse two successful entanglement links. This paper appeals to a more general and efficient swapping method, namely n-fusion based on Greenberger-Horne-Zeilinger measurements that can fuse n successful entanglement links, to maximize the entanglement rate for multiple quantum-user pairs over a quantum network. We propose efficient entanglement routing algorithms that utilize the properties of n-fusion for quantum networks with general topologies. Evaluation results highlight that our proposed algorithm under n-fusion can greatly improve the network performance compared with existing ones.
BinaryViT: Pushing Binary Vision Transformers Towards Convolutional Models
Authors: Phuoc-Hoan Charles Le, Xinlin Li
Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
Abstract
With the increasing popularity and the increasing size of vision transformers (ViTs), there has been an increasing interest in making them more efficient and less computationally costly for deployment on edge devices with limited computing resources. Binarization can be used to help reduce the size of ViT models and their computational cost significantly, using popcount operations when the weights and the activations are in binary. However, ViTs suffer a larger performance drop when directly applying convolutional neural network (CNN) binarization methods or existing binarization methods to binarize ViTs compared to CNNs on datasets with a large number of classes such as ImageNet-1k. With extensive analysis, we find that binary vanilla ViTs such as DeiT miss out on a lot of key architectural properties that CNNs have that allow binary CNNs to have much higher representational capability than binary vanilla ViT. Therefore, we propose BinaryViT, in which inspired by the CNN architecture, we include operations from the CNN architecture into a pure ViT architecture to enrich the representational capability of a binary ViT without introducing convolutions. These include an average pooling layer instead of a token pooling layer, a block that contains multiple average pooling branches, an affine transformation right before the addition of each main residual connection, and a pyramid structure. Experimental results on the ImageNet-1k dataset show the effectiveness of these operations that allow a binary pure ViT model to be competitive with previous state-of-the-art (SOTA) binary CNN models.
Exploring the Representation Power of SPLADE Models
Authors: Joel Mackenzie, Shengyao Zhuang, Guido Zuccon
Abstract
The SPLADE (SParse Lexical AnD Expansion) model is a highly effective approach to learned sparse retrieval, where documents are represented by term impact scores derived from large language models. During training, SPLADE applies regularization to ensure postings lists are kept sparse -- with the aim of mimicking the properties of natural term distributions -- allowing efficient and effective lexical matching and ranking. However, we hypothesize that SPLADE may encode additional signals into common postings lists to further improve effectiveness. To explore this idea, we perform a number of empirical analyses where we re-train SPLADE with different, controlled vocabularies and measure how effective it is at ranking passages. Our findings suggest that SPLADE can effectively encode useful ranking signals in documents even when the vocabulary is constrained to terms that are not traditionally useful for ranking, such as stopwords or even random words.
Decomposing spiking neural networks with Graphical Neural Activity Threads
Authors: Bradley H. Theilman, Felix Wang, Fred Rothganger, James B. Aimone
Subjects: Neural and Evolutionary Computing (cs.NE); Neurons and Cognition (q-bio.NC)
Abstract
A satisfactory understanding of information processing in spiking neural networks requires appropriate computational abstractions of neural activity. Traditionally, the neural population state vector has been the most common abstraction applied to spiking neural networks, but this requires artificially partitioning time into bins that are not obviously relevant to the network itself. We introduce a distinct set of techniques for analyzing spiking neural networks that decomposes neural activity into multiple, disjoint, parallel threads of activity. We construct these threads by estimating the degree of causal relatedness between pairs of spikes, then use these estimates to construct a directed acyclic graph that traces how the network activity evolves through individual spikes. We find that this graph of spiking activity naturally decomposes into disjoint connected components that overlap in space and time, which we call Graphical Neural Activity Threads (GNATs). We provide an efficient algorithm for finding analogous threads that reoccur in large spiking datasets, revealing that seemingly distinct spike trains are composed of similar underlying threads of activity, a hallmark of compositionality. The picture of spiking neural networks provided by our GNAT analysis points to new abstractions for spiking neural computation that are naturally adapted to the spatiotemporally distributed dynamics of spiking neural networks.
Improving Time and Memory Efficiency of Genetic Algorithms by Storing Populations as Minimum Spanning Trees of Patches
Authors: Maxim Buzdalov
Subjects: Neural and Evolutionary Computing (cs.NE)
Abstract
In many applications of evolutionary algorithms the computational cost of applying operators and storing populations is comparable to the cost of fitness evaluation. Furthermore, by knowing what exactly has changed in an individual by an operator, it is possible to recompute fitness value much more efficiently than from scratch. The associated time and memory improvements have been available for simple evolutionary algorithms, few specific genetic algorithms and in the context of gray-box optimization, but not for all algorithms, and the main reason is that it is difficult to achieve in algorithms using large arbitrarily structured populations. This paper makes a first step towards improving this situation. We show that storing the population as a minimum spanning tree, where vertices correspond to individuals but only contain meta-information about them, and edges store structural differences, or patches, between the individuals, is a viable alternative to the straightforward implementation. Our experiments suggest that significant, even asymptotic, improvements -- including execution of crossover operators! -- can be achieved in terms of both memory usage and computational costs.
SRL: Scaling Distributed Reinforcement Learning to Over Ten Thousand Cores
Abstract
The ever-growing complexity of reinforcement learning (RL) tasks demands a distributed RL system to efficiently generate and process a massive amount of data to train intelligent agents. However, existing open-source libraries suffer from various limitations, which impede their practical use in challenging scenarios where large-scale training is necessary. While industrial systems from OpenAI and DeepMind have achieved successful large-scale RL training, their system architecture and implementation details remain undisclosed to the community. In this paper, we present a novel abstraction on the dataflows of RL training, which unifies practical RL training across diverse applications into a general framework and enables fine-grained optimizations. Following this abstraction, we develop a scalable, efficient, and extensible distributed RL system called ReaLly Scalable RL (SRL). The system architecture of SRL separates major RL computation components and allows massively parallelized training. Moreover, SRL offers user-friendly and extensible interfaces for customized algorithms. Our evaluation shows that SRL outperforms existing academic libraries in both a single machine and a medium-sized cluster. In a large-scale cluster, the novel architecture of SRL leads to up to 3.7x speedup compared to the design choices adopted by the existing libraries. We also conduct a direct benchmark comparison to OpenAI's industrial system, Rapid, in the challenging hide-and-seek environment. SRL reproduces the same solution as reported by OpenAI with up to 5x speedup in wall-clock time. Furthermore, we also examine the performance of SRL in a much harder variant of the hide-and-seek environment and achieve substantial learning speedup by scaling SRL to over 15k CPU cores and 32 A100 GPUs. Notably, SRL is the first in the academic community to perform RL experiments at such a large scale.
Neural Polarizer: A Lightweight and Effective Backdoor Defense via Purifying Poisoned Features
Abstract
Recent studies have demonstrated the susceptibility of deep neural networks to backdoor attacks. Given a backdoored model, its prediction of a poisoned sample with trigger will be dominated by the trigger information, though trigger information and benign information coexist. Inspired by the mechanism of the optical polarizer that a polarizer could pass light waves with particular polarizations while filtering light waves with other polarizations, we propose a novel backdoor defense method by inserting a learnable neural polarizer into the backdoored model as an intermediate layer, in order to purify the poisoned sample via filtering trigger information while maintaining benign information. The neural polarizer is instantiated as one lightweight linear transformation layer, which is learned through solving a well designed bi-level optimization problem, based on a limited clean dataset. Compared to other fine-tuning-based defense methods which often adjust all parameters of the backdoored model, the proposed method only needs to learn one additional layer, such that it is more efficient and requires less clean data. Extensive experiments demonstrate the effectiveness and efficiency of our method in removing backdoors across various neural network architectures and datasets, especially in the case of very limited clean data.
Dynamic-Resolution Model Learning for Object Pile Manipulation
Abstract
Dynamics models learned from visual observations have shown to be effective in various robotic manipulation tasks. One of the key questions for learning such dynamics models is what scene representation to use. Prior works typically assume representation at a fixed dimension or resolution, which may be inefficient for simple tasks and ineffective for more complicated tasks. In this work, we investigate how to learn dynamic and adaptive representations at different levels of abstraction to achieve the optimal trade-off between efficiency and effectiveness. Specifically, we construct dynamic-resolution particle representations of the environment and learn a unified dynamics model using graph neural networks (GNNs) that allows continuous selection of the abstraction level. During test time, the agent can adaptively determine the optimal resolution at each model-predictive control (MPC) step. We evaluate our method in object pile manipulation, a task we commonly encounter in cooking, agriculture, manufacturing, and pharmaceutical applications. Through comprehensive evaluations both in the simulation and the real world, we show that our method achieves significantly better performance than state-of-the-art fixed-resolution baselines at the gathering, sorting, and redistribution of granular object piles made with various instances like coffee beans, almonds, corn, etc.
IoT Data Processing for Smart City and Semantic Web Applications
Abstract
The world has been experiencing rapid urbanization over the last few decades, putting a strain on existing city infrastructure such as waste management, water supply management, public transport and electricity consumption. We are also seeing increasing pollution levels in cities threatening the environment, natural resources and health conditions. However, we must realize that the real growth lies in urbanization as it provides many opportunities to individuals for better employment, healthcare and better education. However, it is imperative to limit the ill effects of rapid urbanization through integrated action plans to enable the development of growing cities. This gave rise to the concept of a smart city in which all available information associated with a city will be utilized systematically for better city management. The proposed system architecture is divided in subsystems and is discussed in individual chapters. The first chapter introduces and gives overview to the reader of the complete system architecture. The second chapter discusses the data monitoring system and data lake system based on the oneM2M standards. DMS employs oneM2M as a middleware layer to achieve interoperability, and DLS uses a multi-tenant architecture with multiple logical databases, enabling efficient and reliable data management. The third chapter discusses energy monitoring and electric vehicle charging systems developed to illustrate the applicability of the oneM2M standards. The fourth chapter discusses the Data Exchange System based on the Indian Urban Data Exchange framework. DES uses IUDX standard data schema and open APIs to avoid data silos and enable secure data sharing. The fifth chapter discusses the 5D-IoT framework that provides uniform data quality assessment of sensor data with meaningful data descriptions.
Towards Optimal Randomized Strategies in Adversarial Example Game
Abstract
The vulnerability of deep neural network models to adversarial example attacks is a practical challenge in many artificial intelligence applications. A recent line of work shows that the use of randomization in adversarial training is the key to find optimal strategies against adversarial example attacks. However, in a fully randomized setting where both the defender and the attacker can use randomized strategies, there are no efficient algorithm for finding such an optimal strategy. To fill the gap, we propose the first algorithm of its kind, called FRAT, which models the problem with a new infinite-dimensional continuous-time flow on probability distribution spaces. FRAT maintains a lightweight mixture of models for the defender, with flexibility to efficiently update mixing weights and model parameters at each iteration. Furthermore, FRAT utilizes lightweight sampling subroutines to construct a random strategy for the attacker. We prove that the continuous-time limit of FRAT converges to a mixed Nash equilibria in a zero-sum game formed by a defender and an attacker. Experimental results also demonstrate the efficiency of FRAT on CIFAR-10 and CIFAR-100 datasets.
Eigensubspace of Temporal-Difference Dynamics and How It Improves Value Approximation in Reinforcement Learning
Abstract
We propose a novel value approximation method, namely Eigensubspace Regularized Critic (ERC) for deep reinforcement learning (RL). ERC is motivated by an analysis of the dynamics of Q-value approximation error in the Temporal-Difference (TD) method, which follows a path defined by the 1-eigensubspace of the transition kernel associated with the Markov Decision Process (MDP). It reveals a fundamental property of TD learning that has remained unused in previous deep RL approaches. In ERC, we propose a regularizer that guides the approximation error tending towards the 1-eigensubspace, resulting in a more efficient and stable path of value approximation. Moreover, we theoretically prove the convergence of the ERC method. Besides, theoretical analysis and experiments demonstrate that ERC effectively reduces the variance of value functions. Among 26 tasks in the DMControl benchmark, ERC outperforms state-of-the-art methods for 20. Besides, it shows significant advantages in Q-value approximation and variance reduction. Our code is available at https://sites.google.com/view/erc-ecml23/.
Unified Language Representation for Question Answering over Text, Tables, and Images
Abstract
When trying to answer complex questions, people often rely on multiple sources of information, such as visual, textual, and tabular data. Previous approaches to this problem have focused on designing input features or model structure in the multi-modal space, which is inflexible for cross-modal reasoning or data-efficient training. In this paper, we call for an alternative paradigm, which transforms the images and tables into unified language representations, so that we can simplify the task into a simpler textual QA problem that can be solved using three steps: retrieval, ranking, and generation, all within a language space. This idea takes advantage of the power of pre-trained language models and is implemented in a framework called Solar. Our experimental results show that Solar outperforms all existing methods by 10.6-32.3 pts on two datasets, MultimodalQA and MMCoQA, across ten different metrics. Additionally, Solar achieves the best performance on the WebQA leaderboard
Stop Pre-Training: Adapt Visual-Language Models to Unseen Languages
Authors: Yasmine Karoui, Rémi Lebret, Negar Foroutan, Karl Aberer
Abstract
Vision-Language Pre-training (VLP) has advanced the performance of many vision-language tasks, such as image-text retrieval, visual entailment, and visual reasoning. The pre-training mostly utilizes lexical databases and image queries in English. Previous work has demonstrated that the pre-training in English does not transfer well to other languages in a zero-shot setting. However, multilingual pre-trained language models (MPLM) have excelled at a variety of single-modal language tasks. In this paper, we propose a simple yet efficient approach to adapt VLP to unseen languages using MPLM. We utilize a cross-lingual contextualized token embeddings alignment approach to train text encoders for non-English languages. Our approach does not require image input and primarily uses machine translation, eliminating the need for target language data. Our evaluation across three distinct tasks (image-text retrieval, visual entailment, and natural language visual reasoning) demonstrates that this approach outperforms the state-of-the-art multilingual vision-language models without requiring large parallel corpora. Our code is available at https://github.com/Yasminekaroui/CliCoTea.
Algorithms for Computing Maximum Cliques in Hyperbolic Random Graphs
Abstract
In this paper, we study the maximum clique problem on hyperbolic random graphs. A hyperbolic random graph is a mathematical model for analyzing scale-free networks since it effectively explains the power-law degree distribution of scale-free networks. We propose a simple algorithm for finding a maximum clique in hyperbolic random graph. We first analyze the running time of our algorithm theoretically. We can compute a maximum clique on a hyperbolic random graph $G$ in $O(m + n^{4.5(1-\alpha)})$ expected time if a geometric representation is given or in $O(m + n^{6(1-\alpha)})$ expected time if a geometric representation is not given, where $n$ and $m$ denote the numbers of vertices and edges of $G$, respectively, and $\alpha$ denotes a parameter controlling the power-law exponent of the degree distribution of $G$. Also, we implemented and evaluated our algorithm empirically. Our algorithm outperforms the previous algorithm [BFK18] practically and theoretically. Beyond the hyperbolic random graphs, we have experiment on real-world networks. For most of instances, we get large cliques close to the optimum solutions efficiently.
TacMMs: Tactile Mobile Manipulators for Warehouse Automation
Authors: Zhuochao He, Xuyang Zhang, Simon Jones, Sabine Hauert, Dandan Zhang, Nathan F. Lepora
Abstract
Multi-robot platforms are playing an increasingly important role in warehouse automation for efficient goods transport. This paper proposes a novel customization of a multi-robot system, called Tactile Mobile Manipulators (TacMMs). Each TacMM integrates a soft optical tactile sensor and a mobile robot with a load-lifting mechanism, enabling cooperative transportation in tasks requiring coordinated physical interaction. More specifically, we mount the TacTip (biomimetic optical tactile sensor) on the Distributed Organisation and Transport System (DOTS) mobile robot. The tactile information then helps the mobile robots adjust the relative robot-object pose, thereby increasing the efficiency of load-lifting tasks. This study compares the performance of using two TacMMs with tactile perception with traditional vision-based pose adjustment for load-lifting. The results show that the average success rate of the TacMMs (66%) is improved over a purely visual-based method (34%), with a larger improvement when the mass of the load was non-uniformly distributed. Although this initial study considers two TacMMs, we expect the benefits of tactile perception to extend to multiple mobile robots. Website: https://sites.google.com/view/tacmms
Would I have gotten that reward? Long-term credit assignment by counterfactual contribution analysis
Authors: Alexander Meulemans, Simon Schug, Seijin Kobayashi, Nathaniel Daw, Gregory Wayne
Abstract
To make reinforcement learning more sample efficient, we need better credit assignment methods that measure an action's influence on future rewards. Building upon Hindsight Credit Assignment (HCA), we introduce Counterfactual Contribution Analysis (COCOA), a new family of model-based credit assignment algorithms. Our algorithms achieve precise credit assignment by measuring the contribution of actions upon obtaining subsequent rewards, by quantifying a counterfactual query: "Would the agent still have reached this reward if it had taken another action?". We show that measuring contributions w.r.t. rewarding states, as is done in HCA, results in spurious estimates of contributions, causing HCA to degrade towards the high-variance REINFORCE estimator in many relevant environments. Instead, we measure contributions w.r.t. rewards or learned representations of the rewarding objects, resulting in gradient estimates with lower variance. We run experiments on a suite of problems specifically designed to evaluate long-term credit assignment capabilities. By using dynamic programming, we measure ground-truth policy gradients and show that the improved performance of our new model-based credit assignment methods is due to lower bias and variance compared to HCA and common baselines. Our results demonstrate how modeling action contributions towards rewarding outcomes can be leveraged for credit assignment, opening a new path towards sample-efficient reinforcement learning.
Efficient Sobolev approximation of linear parabolic PDEs in high dimensions
Abstract
In this paper, we study the error in first order Sobolev norm in the approximation of solutions to linear parabolic PDEs. We use a Monte Carlo Euler scheme obtained from combining the Feynman--Kac representation with a Euler discretization of the underlying stochastic process. We derive approximation rates depending on the time-discretization, the number of Monte Carlo simulations, and the dimension. In particular, we show that the Monte Carlo Euler scheme breaks the curse of dimensionality with respect to the first order Sobolev norm. Our argument is based on new estimates on the weak error of the Euler approximation of a diffusion process together with its derivative with respect to the initial condition. As a consequence, we obtain that neural networks are able to approximate solutions of linear parabolic PDEs in first order Sobolev norm without the curse of dimensionality if the coefficients of the PDEs admit an efficient approximation with neural networks.
Length of Stay prediction for Hospital Management using Domain Adaptation
Authors: Lyse Naomi Wamba Momo, Nyalleng Moorosi, Elaine O. Nsoesie, Frank Rademakers, Bart De Moor
Abstract
Inpatient length of stay (LoS) is an important managerial metric which if known in advance can be used to efficiently plan admissions, allocate resources and improve care. Using historical patient data and machine learning techniques, LoS prediction models can be developed. Ethically, these models can not be used for patient discharge in lieu of unit heads but are of utmost necessity for hospital management systems in charge of effective hospital planning. Therefore, the design of the prediction system should be adapted to work in a true hospital setting. In this study, we predict early hospital LoS at the granular level of admission units by applying domain adaptation to leverage information learned from a potential source domain. Time-varying data from 110,079 and 60,492 patient stays to 8 and 9 intensive care units were respectively extracted from eICU-CRD and MIMIC-IV. These were fed into a Long-Short Term Memory and a Fully connected network to train a source domain model, the weights of which were transferred either partially or fully to initiate training in target domains. Shapley Additive exPlanations (SHAP) algorithms were used to study the effect of weight transfer on model explanability. Compared to the benchmark, the proposed weight transfer model showed statistically significant gains in prediction accuracy (between 1% and 5%) as well as computation time (up to 2hrs) for some target domains. The proposed method thus provides an adapted clinical decision support system for hospital management that can ease processes of data access via ethical committee, computation infrastructures and time.
An Exact Characterisation of Flexibility in Populations of Electric Vehicles
Abstract
Increasing penetrations of electric vehicles (EVs) presents a large source of flexibility, which can be used to assist balancing the power grid. The flexibility of an individual EV can be quantified as a convex polytope and the flexibility of a population of EVs is the Minkowski sum of these polytopes. In general computing the exact Minkowski sum is intractable. However, exploiting symmetry in a restricted but significant case, enables an efficient computation of the aggregate flexibility. This results in a polytope with exponentially many vertices and facets with respect to the time horizon. We show how to use a lifting procedure to provide a representation of this polytope with a reduced number of facets, which makes optimising over more tractable. Finally, a disaggregation procedure that takes an aggregate signal and computes dispatch instructions for each EV in the population is presented. The complexity of the algorithms presented is independent of the size of the population and polynomial in the length of the time horizon. We evaluate this work against existing methods in the literature, and show how this method guarantees optimality with lower computational burden than existing methods.
Sampling weights of deep neural networks
Authors: Erik Lien Bolager, Iryna Burak, Chinmay Datar, Qing Sun, Felix Dietrich
Abstract
We introduce a probability distribution, combined with an efficient sampling algorithm, for weights and biases of fully-connected neural networks. In a supervised learning context, no iterative optimization or gradient computations of internal network parameters are needed to obtain a trained network. The sampling is based on the idea of random feature models. However, instead of a data-agnostic distribution, e.g., a normal distribution, we use both the input and the output training data of the supervised learning problem to sample both shallow and deep networks. We prove that the sampled networks we construct are universal approximators. We also show that our sampling scheme is invariant to rigid body transformations and scaling of the input data. This implies many popular pre-processing techniques are no longer required. For Barron functions, we show that the $L^2$-approximation error of sampled shallow networks decreases with the square root of the number of neurons. In numerical experiments, we demonstrate that sampled networks achieve comparable accuracy as iteratively trained ones, but can be constructed orders of magnitude faster. Our test cases involve a classification benchmark from OpenML, sampling of neural operators to represent maps in function spaces, and transfer learning using well-known architectures.
Tokenization and the Noiseless Channel
Authors: Vilém Zouhar, Clara Meister, Juan Luis Gastaldi, Li Du, Mrinmaya Sachan, Ryan Cotterell
Subjects: Computation and Language (cs.CL); Information Theory (cs.IT)
Abstract
Subword tokenization is a key part of many NLP pipelines. However, little is known about why some tokenizer and hyperparameter combinations lead to better downstream model performance than others. We propose that good tokenizers lead to \emph{efficient} channel usage, where the channel is the means by which some input is conveyed to the model and efficiency can be quantified in information-theoretic terms as the ratio of the Shannon entropy to the maximum possible entropy of the token distribution. Yet, an optimal encoding according to Shannon entropy assigns extremely long codes to low-frequency tokens and very short codes to high-frequency tokens. Defining efficiency in terms of R\'enyi entropy, on the other hand, penalizes distributions with either very high or very low-frequency tokens. In machine translation, we find that across multiple tokenizers, the R\'enyi entropy with $\alpha = 2.5$ has a very strong correlation with \textsc{Bleu}: $0.78$ in comparison to just $-0.32$ for compressed length.
ICDaeLST: Intensity-Controllable Detail Attention-enhanced for Lightweight Fast Style Transfer
Authors: Jiang Shi Qi
Subjects: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
Abstract
The mainstream style transfer methods usually use pre-trained deep convolutional neural network (VGG) models as encoders, or use more complex model structures to achieve better style transfer effects. This leads to extremely slow processing speeds for practical tasks due to limited resources or higher resolution image processing, such as 4K images, severely hindering the practical application value of style transfer models. We introduce a lightweight and fast styletransfer model with controllable detail attention enhancement, named ICDaeLST. The model adopts a minimal, shallow, and small architecture, forming a very compact lightweight model for efficient forward inference. Although its structure is simple and has limited parameters, we achieve better overall color and texture structure matching by introducing a style discriminator, design additional global semantic invariance loss to preserve the semantic and structural information of the content image from a high-level global perspective, and design a shallow detail attention enhancement module to preserve the detail information of the content image from a low-level detail perspective. We also achieve controllable intensity during inference for the first time (adjusting the degree of detail retention and texture structure transfer based on subjective judgment) to meet different users' subjective evaluation of stylization effects. Compared with the current best-performing and most lightweight models, our model achieves better style transfer quality and better content structure and detail retention, while having a smaller model size (17-250 times smaller) and faster speed (0.26-6.5 times faster), and achieves the fastest processing speed of 0.38s on 4K high-resolution images.
Abstract
Shifting social opinions has far-reaching implications in various aspects, such as public health campaigns, product marketing, and political candidates. In this paper, we study a problem of opinion optimization based on the popular Friedkin-Johnsen (FJ) model for opinion dynamics in an unweighted directed social network with $n$ nodes and $m$ edges. In the FJ model, the internal opinion of every node lies in the closed interval $[0, 1]$, with 0 and 1 being polar opposites of opinions about a certain issue. Concretely, we focus on the problem of selecting a small number of $ k\ll n $ nodes and changing their internal opinions to 0, in order to minimize the average opinion at equilibrium. We then design an algorithm that returns the optimal solution to the problem in $O(n^3)$ time. To speed up the computation, we further develop a fast algorithm by sampling spanning forests, the time complexity of which is $ O(ln) $, with $l$ being the number of samplings. Finally, we execute extensive experiments on various real directed networks, which show that the effectiveness of our two algorithms is similar to each other, both of which outperform several baseline strategies of node selection. Moreover, our fast algorithm is more efficient than the first one, which is scalable to massive graphs with more than twenty million nodes.
SWAT: A System-Wide Approach to Tunable Leakage Mitigation in Encrypted Data Stores
Authors: Leqian Zheng, Lei Xu, Cong Wang, Sheng Wang, Yuke Hu, Zhan Qin, Feifei Li, Kui Ren
Abstract
Numerous studies have underscored the significant privacy risks associated with various leakage patterns in encrypted data stores. Most existing systems that conceal leakage either (1) incur substantial overheads, (2) focus on specific subsets of leakage patterns, or (3) apply the same security notion across various workloads, thereby impeding the attainment of fine-tuned privacy-efficiency trade-offs. In light of various detrimental leakage patterns, this paper starts with an investigation into which specific leakage patterns require our focus respectively in the contexts of key-value, range-query, and dynamic workloads. Subsequently, we introduce new security notions tailored to the specific privacy requirements of these workloads. Accordingly, we present, SWAT, an efficient construction that progressively enables these workloads, while provably mitigating system-wide leakage via a suite of algorithms with tunable privacy-efficiency trade-offs. We conducted extensive experiments and compiled a detailed result analysis, showing the efficiency of our solution. SWAT is about $10.6\times$ slower than an encryption-only data store that reveals various leakage patterns and is $31.6\times$ faster than a trivially zero-leakage solution. Meanwhile, the performance of SWAT remains highly competitive compared to other designs that mitigate specific types of leakage.
Sustainable Palm Tree Farming: Leveraging IoT and Multi-Modal Data for Early Detection and Mapping of Red Palm Weevil
Abstract
The Red Palm Weevil (RPW) is a highly destructive insect causing economic losses and impacting palm tree farming worldwide. This paper proposes an innovative approach for sustainable palm tree farming by utilizing advanced technologies for the early detection and management of RPW. Our approach combines computer vision, deep learning (DL), the Internet of Things (IoT), and geospatial data to detect and classify RPW-infested palm trees effectively. The main phases include; (1) DL classification using sound data from IoT devices, (2) palm tree detection using YOLOv8 on UAV images, and (3) RPW mapping using geospatial data. Our custom DL model achieves 100% precision and recall in detecting and localizing infested palm trees. Integrating geospatial data enables the creation of a comprehensive RPW distribution map for efficient monitoring and targeted management strategies. This technology-driven approach benefits agricultural authorities, farmers, and researchers in managing RPW infestations and safeguarding palm tree plantations' productivity.
The ELM Neuron: an Efficient and Expressive Cortical Neuron Model Can Solve Long-Horizon Tasks
Authors: Aaron Spieler, Nasim Rahaman, Georg Martius, Bernhard Schölkopf, Anna Levina
Subjects: Neural and Evolutionary Computing (cs.NE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Neurons and Cognition (q-bio.NC)
Abstract
Traditional large-scale neuroscience models and machine learning utilize simplified models of individual neurons, relying on collective activity and properly adjusted connections to perform complex computations. However, each biological cortical neuron is inherently a sophisticated computational device, as corroborated in a recent study where it took a deep artificial neural network with millions of parameters to replicate the input-output relationship of a detailed biophysical model of a cortical pyramidal neuron. We question the necessity for these many parameters and introduce the Expressive Leaky Memory (ELM) neuron, a biologically inspired, computationally expressive, yet efficient model of a cortical neuron. Remarkably, our ELM neuron requires only 8K trainable parameters to match the aforementioned input-output relationship accurately. We find that an accurate model necessitates multiple memory-like hidden states and intricate nonlinear synaptic integration. To assess the computational ramifications of this design, we evaluate the ELM neuron on various tasks with demanding temporal structures, including a sequential version of the CIFAR-10 classification task, the challenging Pathfinder-X task, and a new dataset based on the Spiking Heidelberg Digits dataset. Our ELM neuron outperforms most transformer-based models on the Pathfinder-X task with 77% accuracy, demonstrates competitive performance on Sequential CIFAR-10, and superior performance compared to classic LSTM models on the variant of the Spiking Heidelberg Digits dataset. These findings indicate a potential for biologically motivated, computationally efficient neuronal models to enhance performance in challenging machine learning tasks.
MIS-FM: 3D Medical Image Segmentation using Foundation Models Pretrained on a Large-Scale Unannotated Dataset
Abstract
Pretraining with large-scale 3D volumes has a potential for improving the segmentation performance on a target medical image dataset where the training images and annotations are limited. Due to the high cost of acquiring pixel-level segmentation annotations on the large-scale pretraining dataset, pretraining with unannotated images is highly desirable. In this work, we propose a novel self-supervised learning strategy named Volume Fusion (VF) for pretraining 3D segmentation models. It fuses several random patches from a foreground sub-volume to a background sub-volume based on a predefined set of discrete fusion coefficients, and forces the model to predict the fusion coefficient of each voxel, which is formulated as a self-supervised segmentation task without manual annotations. Additionally, we propose a novel network architecture based on parallel convolution and transformer blocks that is suitable to be transferred to different downstream segmentation tasks with various scales of organs and lesions. The proposed model was pretrained with 110k unannotated 3D CT volumes, and experiments with different downstream segmentation targets including head and neck organs, thoracic/abdominal organs showed that our pretrained model largely outperformed training from scratch and several state-of-the-art self-supervised training methods and segmentation models. The code and pretrained model are available at https://github.com/openmedlab/MIS-FM.
OSP: Boosting Distributed Model Training with 2-stage Synchronization
Authors: Zixuan Chen, Lei Shi, Xuandong Liu, Jiahui Li, Sen Liu, Yang Xu
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC); Machine Learning (cs.LG)
Abstract
Distributed deep learning (DDL) is a promising research area, which aims to increase the efficiency of training deep learning tasks with large size of datasets and models. As the computation capability of DDL nodes continues to increase, the network connection between nodes is becoming a major bottleneck. Various methods of gradient compression and improved model synchronization have been proposed to address this bottleneck in Parameter-Server-based DDL. However, these two types of methods can result in accuracy loss due to discarded gradients and have limited enhancement on the throughput of model synchronization, respectively. To address these challenges, we propose a new model synchronization method named Overlapped Synchronization Parallel (OSP), which achieves efficient communication with a 2-stage synchronization approach and uses Local-Gradient-based Parameter correction (LGP) to avoid accuracy loss caused by stale parameters. The prototype of OSP has been implemented using PyTorch and evaluated on commonly used deep learning models and datasets with a 9-node testbed. Evaluation results show that OSP can achieve up to 50\% improvement in throughput without accuracy loss compared to popular synchronization models.
The War of the Efficiencies: Understanding the Tension between Carbon and Energy Optimization
Authors: Walid A. Hanafy, Roozbeh Bostandoost, Noman Bashir, David Irwin, Mohammad Hajiesmaili, Prashant Shenoy
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract
Major innovations in computing have been driven by scaling up computing infrastructure, while aggressively optimizing operating costs. The result is a network of worldwide datacenters that consume a large amount of energy, mostly in an energy-efficient manner. Since the electric grid powering these datacenters provided a simple and opaque abstraction of an unlimited and reliable power supply, the computing industry remained largely oblivious to the carbon intensity of the electricity it uses. Much like the rest of the society, it generally treated the carbon intensity of the electricity as constant, which was mostly true for a fossil fuel-driven grid. As a result, the cost-driven objective of increasing energy-efficiency -- by doing more work per unit of energy -- has generally been viewed as the most carbon-efficient approach. However, as the electric grid is increasingly powered by clean energy and is exposing its time-varying carbon intensity, the most energy-efficient operation is no longer necessarily the most carbon-efficient operation. There has been a recent focus on exploiting the flexibility of computing's workloads -- along temporal, spatial, and resource dimensions -- to reduce carbon emissions, which comes at the cost of either performance or energy efficiency. In this paper, we discuss the trade-offs between energy efficiency and carbon efficiency in exploiting computing's flexibility and show that blindly optimizing for energy efficiency is not always the right approach.
Experience Transfer for Robust Direct Data-Driven Control
Authors: Alexander von Rohr, Dmitrii Likhachev, Sebastian Trimpe
Subjects: Robotics (cs.RO); Systems and Control (eess.SY)
Abstract
Learning-based control uses data to design efficient controllers for specific systems. When multiple systems are involved, experience transfer usually focuses on data availability and controller performance yet neglects robustness to variations between systems. In contrast, this letter explores experience transfer from a robustness perspective. We leverage the transfer to design controllers that are robust not only to the uncertainty regarding an individual agent's model but also to the choice of agent in a fleet. Experience transfer enables the design of safe and robust controllers that work out of the box for all systems in a heterogeneous fleet. Our approach combines scenario optimization and recent formulations for direct data-driven control without the need to estimate a model of the system or determine uncertainty bounds for its parameters. We demonstrate the benefits of our data-driven robustification method through a numerical case study and obtain learned controllers that generalize well from a small number of open-loop trajectories in a quadcopter simulation.
Diffusion-Jump GNNs: Homophiliation via Learnable Metric Filters
Authors: Ahmed Begga, Francisco Escolano, Miguel Angel Lozano, Edwin R. Hancock
Abstract
High-order Graph Neural Networks (HO-GNNs) have been developed to infer consistent latent spaces in the heterophilic regime, where the label distribution is not correlated with the graph structure. However, most of the existing HO-GNNs are hop-based, i.e., they rely on the powers of the transition matrix. As a result, these architectures are not fully reactive to the classification loss and the achieved structural filters have static supports. In other words, neither the filters' supports nor their coefficients can be learned with these networks. They are confined, instead, to learn combinations of filters. To address the above concerns, we propose Diffusion-jump GNNs a method relying on asymptotic diffusion distances that operates on jumps. A diffusion-pump generates pairwise distances whose projections determine both the support and coefficients of each structural filter. These filters are called jumps because they explore a wide range of scales in order to find bonds between scattered nodes with the same label. Actually, the full process is controlled by the classification loss. Both the jumps and the diffusion distances react to classification errors (i.e. they are learnable). Homophiliation, i.e., the process of learning piecewise smooth latent spaces in the heterophilic regime, is formulated as a Dirichlet problem: the known labels determine the border nodes and the diffusion-pump ensures a minimal deviation of the semi-supervised grouping from a canonical unsupervised grouping. This triggers the update of both the diffusion distances and, consequently, the jumps in order to minimize the classification error. The Dirichlet formulation has several advantages. It leads to the definition of structural heterophily, a novel measure beyond edge heterophily. It also allows us to investigate links with (learnable) diffusion distances, absorbing random walks and stochastic diffusion.
Integrating Large Pre-trained Models into Multimodal Named Entity Recognition with Evidential Fusion
Abstract
Multimodal Named Entity Recognition (MNER) is a crucial task for information extraction from social media platforms such as Twitter. Most current methods rely on attention weights to extract information from both text and images but are often unreliable and lack interpretability. To address this problem, we propose incorporating uncertainty estimation into the MNER task, producing trustworthy predictions. Our proposed algorithm models the distribution of each modality as a Normal-inverse Gamma distribution, and fuses them into a unified distribution with an evidential fusion mechanism, enabling hierarchical characterization of uncertainties and promotion of prediction accuracy and trustworthiness. Additionally, we explore the potential of pre-trained large foundation models in MNER and propose an efficient fusion approach that leverages their robust feature representations. Experiments on two datasets demonstrate that our proposed method outperforms the baselines and achieves new state-of-the-art performance.
Spatial Reasoning via Deep Vision Models for Robotic Sequential Manipulation
Authors: Hongyou Zhou, Ingmar Fabian Schubert, Marc Toussaint, Ozgur S. Oguz
Abstract
In this paper, we propose using deep neural architectures (i.e., vision transformers and ResNet) as heuristics for sequential decision-making in robotic manipulation problems. This formulation enables predicting the subset of objects that are relevant for completing a task. Such problems are often addressed by task and motion planning (TAMP) formulations combining symbolic reasoning and continuous motion planning. In essence, the action-object relationships are resolved for discrete, symbolic decisions that are used to solve manipulation motions (e.g., via nonlinear trajectory optimization). However, solving long-horizon tasks requires consideration of all possible action-object combinations which limits the scalability of TAMP approaches. To overcome this combinatorial complexity, we introduce a visual perception module integrated with a TAMP-solver. Given a task and an initial image of the scene, the learned model outputs the relevancy of objects to accomplish the task. By incorporating the predictions of the model into a TAMP formulation as a heuristic, the size of the search space is significantly reduced. Results show that our framework finds feasible solutions more efficiently when compared to a state-of-the-art TAMP solver.
Localized implicit time stepping for the wave equation
Abstract
This work proposes a discretization of the acoustic wave equation with possibly oscillatory coefficients based on a superposition of discrete solutions to spatially localized subproblems computed with an implicit time discretization. Based on exponentially decaying entries of the global system matrices and an appropriate partition of unity, it is proved that the superposition of localized solutions is appropriately close to the solution of the (global) implicit scheme. It is thereby justified that the localized (and especially parallel) computation on multiple overlapping subdomains is reasonable. Moreover, a re-start is introduced after a certain amount of time steps to maintain a moderate overlap of the subdomains. Overall, the approach may be understood as a domain decomposition strategy (in space and time) that completely avoids inner iterations. Numerical examples are presented.
On the Predictive Accuracy of Neural Temporal Point Process Models for Continuous-time Event Data
Abstract
Temporal Point Processes (TPPs) serve as the standard mathematical framework for modeling asynchronous event sequences in continuous time. However, classical TPP models are often constrained by strong assumptions, limiting their ability to capture complex real-world event dynamics. To overcome this limitation, researchers have proposed Neural TPPs, which leverage neural network parametrizations to offer more flexible and efficient modeling. While recent studies demonstrate the effectiveness of Neural TPPs, they often lack a unified setup, relying on different baselines, datasets, and experimental configurations. This makes it challenging to identify the key factors driving improvements in predictive accuracy, hindering research progress. To bridge this gap, we present a comprehensive large-scale experimental study that systematically evaluates the predictive accuracy of state-of-the-art neural TPP models. Our study encompasses multiple real-world and synthetic event sequence datasets, following a carefully designed unified setup. We thoroughly investigate the influence of major architectural components such as event encoding, history encoder, and decoder parametrization on both time and mark prediction tasks. Additionally, we delve into the less explored area of probabilistic calibration for neural TPP models. By analyzing our results, we draw insightful conclusions regarding the significance of history size and the impact of architectural components on predictive accuracy. Furthermore, we shed light on the miscalibration of mark distributions in neural TPP models. Our study aims to provide valuable insights into the performance and characteristics of neural TPP models, contributing to a better understanding of their strengths and limitations.
Detect Any Deepfakes: Segment Anything Meets Face Forgery Detection and Localization
Authors: Yingxin Lai, Zhiming Luo, Zitong Yu
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
The rapid advancements in computer vision have stimulated remarkable progress in face forgery techniques, capturing the dedicated attention of researchers committed to detecting forgeries and precisely localizing manipulated areas. Nonetheless, with limited fine-grained pixel-wise supervision labels, deepfake detection models perform unsatisfactorily on precise forgery detection and localization. To address this challenge, we introduce the well-trained vision segmentation foundation model, i.e., Segment Anything Model (SAM) in face forgery detection and localization. Based on SAM, we propose the Detect Any Deepfakes (DADF) framework with the Multiscale Adapter, which can capture short- and long-range forgery contexts for efficient fine-tuning. Moreover, to better identify forged traces and augment the model's sensitivity towards forgery regions, Reconstruction Guided Attention (RGA) module is proposed. The proposed framework seamlessly integrates end-to-end forgery localization and detection optimization. Extensive experiments on three benchmark datasets demonstrate the superiority of our approach for both forgery detection and localization. The codes will be released soon at https://github.com/laiyingxin2/DADF.
RL4CO: an Extensive Reinforcement Learning for Combinatorial Optimization Benchmark
Authors: Federico Berto, Chuanbo Hua, Junyoung Park, Minsu Kim, Hyeonah Kim, Jiwoo Son, Haeyeon Kim, Joungho Kim, Jinkyoo Park
Abstract
We introduce RL4CO, an extensive reinforcement learning (RL) for combinatorial optimization (CO) benchmark. RL4CO employs state-of-the-art software libraries as well as best practices in implementation, such as modularity and configuration management, to be efficient and easily modifiable by researchers for adaptations of neural network architecture, environments, and algorithms. Contrary to the existing focus on specific tasks like the traveling salesman problem (TSP) for performance assessment, we underline the importance of scalability and generalization capabilities for diverse optimization tasks. We also systematically benchmark sample efficiency, zero-shot generalization, and adaptability to changes in data distributions of various models. Our experiments show that some recent state-of-the-art methods fall behind their predecessors when evaluated using these new metrics, suggesting the necessity for a more balanced view of the performance of neural CO solvers. We hope RL4CO will encourage the exploration of novel solutions to complex real-world tasks, allowing to compare with existing methods through a standardized interface that decouples the science from the software engineering. We make our library publicly available at https://github.com/kaist-silab/rl4co.
PVP: Personalized Video Prior for Editable Dynamic Portraits using StyleGAN
Authors: Kai-En Lin, Alex Trevithick, Keli Cheng, Michel Sarkis, Mohsen Ghafoorian, Ning Bi, Gerhard Reitmayr, Ravi Ramamoorthi
Subjects: Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR)
Abstract
Portrait synthesis creates realistic digital avatars which enable users to interact with others in a compelling way. Recent advances in StyleGAN and its extensions have shown promising results in synthesizing photorealistic and accurate reconstruction of human faces. However, previous methods often focus on frontal face synthesis and most methods are not able to handle large head rotations due to the training data distribution of StyleGAN. In this work, our goal is to take as input a monocular video of a face, and create an editable dynamic portrait able to handle extreme head poses. The user can create novel viewpoints, edit the appearance, and animate the face. Our method utilizes pivotal tuning inversion (PTI) to learn a personalized video prior from a monocular video sequence. Then we can input pose and expression coefficients to MLPs and manipulate the latent vectors to synthesize different viewpoints and expressions of the subject. We also propose novel loss functions to further disentangle pose and expression in the latent space. Our algorithm shows much better performance over previous approaches on monocular video datasets, and it is also capable of running in real-time at 54 FPS on an RTX 3080.
Ducho: A Unified Framework for the Extraction of Multimodal Features in Recommendation
Authors: Daniele Malitesta, Giuseppe Gassi, Claudio Pomo, Tommaso Di Noia
Subjects: Information Retrieval (cs.IR); Multimedia (cs.MM)
Abstract
In multimodal-aware recommendation, the extraction of meaningful multimodal features is at the basis of high-quality recommendations. Generally, each recommendation framework implements its multimodal extraction procedures with specific strategies and tools. This is limiting for two reasons: (i) different extraction strategies do not ease the interdependence among multimodal recommendation frameworks; thus, they cannot be efficiently and fairly compared; (ii) given the large plethora of pre-trained deep learning models made available by different open source tools, model designers do not have access to shared interfaces to extract features. Motivated by the outlined aspects, we propose Ducho, a unified framework for the extraction of multimodal features in recommendation. By integrating three widely-adopted deep learning libraries as backends, namely, TensorFlow, PyTorch, and Transformers, we provide a shared interface to extract and process features where each backend's specific methods are abstracted to the end user. Noteworthy, the extraction pipeline is easily configurable with a YAML-based file where the user can specify, for each modality, the list of models (and their specific backends/parameters) to perform the extraction. Finally, to make Ducho accessible to the community, we build a public Docker image equipped with a ready-to-use CUDA environment and propose three demos to test its functionalities for different scenarios and tasks. The GitHub repository and the documentation is accessible at this link: https://github.com/sisinflab/Ducho.
Generate Anything Anywhere in Any Scene
Authors: Yuheng Li, Haotian Liu, Yangming Wen, Yong Jae Lee
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Text-to-image diffusion models have attracted considerable interest due to their wide applicability across diverse fields. However, challenges persist in creating controllable models for personalized object generation. In this paper, we first identify the entanglement issues in existing personalized generative models, and then propose a straightforward and efficient data augmentation training strategy that guides the diffusion model to focus solely on object identity. By inserting the plug-and-play adapter layers from a pre-trained controllable diffusion model, our model obtains the ability to control the location and size of each generated personalized object. During inference, we propose a regionally-guided sampling technique to maintain the quality and fidelity of the generated images. Our method achieves comparable or superior fidelity for personalized objects, yielding a robust, versatile, and controllable text-to-image diffusion model that is capable of generating realistic and personalized images. Our approach demonstrates significant potential for various applications, such as those in art, entertainment, and advertising design.
FogROS2-SGC: A ROS2 Cloud Robotics Platform for Secure Global Connectivity
Authors: Kaiyuan Chen, Ryan Hoque, Karthik Dharmarajan, Edith LLontop, Simeon Adebola, Jeffrey Ichnowski, John Kubiatowicz, Ken Goldberg
Abstract
The Robot Operating System (ROS2) is the most widely used software platform for building robotics applications. FogROS2 extends ROS2 to allow robots to access cloud computing on demand. However, ROS2 and FogROS2 assume that all robots are locally connected and that each robot has full access and control of the other robots. With applications like distributed multi-robot systems, remote robot control, and mobile robots, robotics increasingly involves the global Internet and complex trust management. Existing approaches for connecting disjoint ROS2 networks lack key features such as security, compatibility, efficiency, and ease of use. We introduce FogROS2-SGC, an extension of FogROS2 that can effectively connect robot systems across different physical locations, networks, and Data Distribution Services (DDS). With globally unique and location-independent identifiers, FogROS2-SGC securely and efficiently routes data between robotics components around the globe. FogROS2-SGC is agnostic to the ROS2 distribution and configuration, is compatible with non-ROS2 software, and seamlessly extends existing ROS2 applications without any code modification. Experiments suggest FogROS2-SGC is 19x faster than rosbridge (a ROS2 package with comparable features, but lacking security). We also apply FogROS2-SGC to 4 robots and compute nodes that are 3600km apart. Videos and code are available on the project website https://sites.google.com/view/fogros2-sgc.
An Efficient General-Purpose Modular Vision Model via Multi-Task Heterogeneous Training
Authors: Zitian Chen, Mingyu Ding, Yikang Shen, Wei Zhan, Masayoshi Tomizuka, Erik Learned-Miller, Chuang Gan
Abstract
We present a model that can perform multiple vision tasks and can be adapted to other downstream tasks efficiently. Despite considerable progress in multi-task learning, most efforts focus on learning from multi-label data: a single image set with multiple task labels. Such multi-label data sets are rare, small, and expensive. We say heterogeneous to refer to image sets with different task labels, or to combinations of single-task datasets. Few have explored training on such heterogeneous datasets. General-purpose vision models are still dominated by single-task pretraining, and it remains unclear how to scale up multi-task models by leveraging mainstream vision datasets designed for different purposes. The challenges lie in managing large intrinsic differences among vision tasks, including data distribution, architectures, task-specific modules, dataset scales, and sampling strategies. To address these challenges, we propose to modify and scale up mixture-of-experts (MoE) vision transformers, so that they can simultaneously learn classification, detection, and segmentation on diverse mainstream vision datasets including ImageNet, COCO, and ADE20K. Our approach achieves comparable results to single-task state-of-the-art models and demonstrates strong generalization on downstream tasks. Due to its emergent modularity, this general-purpose model decomposes into high-performing components, efficiently adapting to downstream tasks. We can fine-tune it with fewer training parameters, fewer model parameters, and less computation. Additionally, its modularity allows for easy expansion in continual-learning-without-forgetting scenarios. Finally, these functions can be controlled and combined to meet various demands of downstream tasks.
Abstract
The two-time scale nature of SAC, which is an actor-critic algorithm, is characterised by the fact that the critic estimate has not converged for the actor at any given time, but since the critic learns faster than the actor, it ensures eventual consistency between the two. Various strategies have been introduced in literature to learn better gradient estimates to help achieve better convergence. Since gradient estimates depend upon the critic, we posit that improving the critic can provide a better gradient estimate for the actor at each time. Utilizing this, we propose Soft Actor Retrospective Critic (SARC), where we augment the SAC critic loss with another loss term - retrospective loss - leading to faster critic convergence and consequently, better policy gradient estimates for the actor. An existing implementation of SAC can be easily adapted to SARC with minimal modifications. Through extensive experimentation and analysis, we show that SARC provides consistent improvement over SAC on benchmark environments. We plan to open-source the code and all experiment data at: https://github.com/sukritiverma1996/SARC.
Momentum Benefits Non-IID Federated Learning Simply and Provably
Authors: Ziheng Cheng, Xinmeng Huang, Kun Yuan
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)
Abstract
Federated learning is a powerful paradigm for large-scale machine learning, but it faces significant challenges due to unreliable network connections, slow communication, and substantial data heterogeneity across clients. FedAvg and SCAFFOLD are two fundamental algorithms to address these challenges. In particular, FedAvg employs multiple local updates before communicating with a central server, while SCAFFOLD maintains a control variable on each client to compensate for "client drift" in its local updates. Various methods have been proposed in literature to enhance the convergence of these two algorithms, but they either make impractical adjustments to algorithmic structure, or rely on the assumption of bounded data heterogeneity. This paper explores the utilization of momentum to enhance the performance of FedAvg and SCAFFOLD. When all clients participate in the training process, we demonstrate that incorporating momentum allows FedAvg to converge without relying on the assumption of bounded data heterogeneity even using a constant local learning rate. This is a novel result since existing analyses for FedAvg require bounded data heterogeneity even with diminishing local learning rates. In the case of partial client participation, we show that momentum enables SCAFFOLD to converge provably faster without imposing any additional assumptions. Furthermore, we use momentum to develop new variance-reduced extensions of FedAvg and SCAFFOLD, which exhibit state-of-the-art convergence rates. Our experimental results support all theoretical findings.
A systematic study of the foreground-background imbalance problem in deep learning for object detection
Authors: Hanxue Gu, Haoyu Dong, Nicholas Konz, Maciej A. Mazurowski
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
The class imbalance problem in deep learning has been explored in several studies, but there has yet to be a systematic analysis of this phenomenon in object detection. Here, we present comprehensive analyses and experiments of the foreground-background (F-B) imbalance problem in object detection, which is very common and caused by small, infrequent objects of interest. We experimentally study the effects of different aspects of F-B imbalance (object size, number of objects, dataset size, object type) on detection performance. In addition, we also compare 9 leading methods for addressing this problem, including Faster-RCNN, SSD, OHEM, Libra-RCNN, Focal-Loss, GHM, PISA, YOLO-v3, and GFL with a range of datasets from different imaging domains. We conclude that (1) the F-B imbalance can indeed cause a significant drop in detection performance, (2) The detection performance is more affected by F-B imbalance when fewer training data are available, (3) in most cases, decreasing object size leads to larger performance drop than decreasing number of objects, given the same change in the ratio of object pixels to non-object pixels, (6) among all selected methods, Libra-RCNN and PISA demonstrate the best performance in addressing the issue of F-B imbalance. (7) When the training dataset size is large, the choice of method is not impactful (8) Soft-sampling methods, including focal-loss, GHM, and GFL, perform fairly well on average but are relatively unstable.
Envisioning a Next Generation Extended Reality Conferencing System with Efficient Photorealistic Human Rendering
Abstract
Meeting online is becoming the new normal. Creating an immersive experience for online meetings is a necessity towards more diverse and seamless environments. Efficient photorealistic rendering of human 3D dynamics is the core of immersive meetings. Current popular applications achieve real-time conferencing but fall short in delivering photorealistic human dynamics, either due to limited 2D space or the use of avatars that lack realistic interactions between participants. Recent advances in neural rendering, such as the Neural Radiance Field (NeRF), offer the potential for greater realism in metaverse meetings. However, the slow rendering speed of NeRF poses challenges for real-time conferencing. We envision a pipeline for a future extended reality metaverse conferencing system that leverages monocular video acquisition and free-viewpoint synthesis to enhance data and hardware efficiency. Towards an immersive conferencing experience, we explore an accelerated NeRF-based free-viewpoint synthesis algorithm for rendering photorealistic human dynamics more efficiently. We show that our algorithm achieves comparable rendering quality while performing training and inference 44.5% and 213% faster than state-of-the-art methods, respectively. Our exploration provides a design basis for constructing metaverse conferencing systems that can handle complex application scenarios, including dynamic scene relighting with customized themes and multi-user conferencing that harmonizes real-world people into an extended world.
Learning from Synthetic Human Group Activities
Authors: Che-Jui Chang, Honglu Zhou, Parth Goel, Aditya Bhat, Seonghyeon Moon, Samuel S. Sohn, Sejong Yoon, Vladimir Pavlovic, Mubbasir Kapadia
Abstract
The understanding of complex human interactions and group activities has garnered attention in human-centric computer vision. However, the advancement of the related tasks is hindered due to the difficulty of obtaining large-scale labeled real-world datasets. To mitigate the issue, we propose M3Act, a multi-view multi-group multi-person human atomic action and group activity data generator. Powered by the Unity engine, M3Act contains simulation-ready 3D scenes and human assets, configurable lighting and camera systems, highly parameterized modular group activities, and a large degree of domain randomization during the data generation process. Our data generator is capable of generating large-scale datasets of human activities with multiple viewpoints, modalities (RGB images, 2D poses, 3D motions), and high-quality annotations for individual persons and multi-person groups (2D bounding boxes, instance segmentation masks, individual actions and group activity categories). Using M3Act, we perform synthetic data pre-training for 2D skeleton-based group activity recognition and RGB-based multi-person pose tracking. The results indicate that learning from our synthetic datasets largely improves the model performances on real-world datasets, with the highest gain of 5.59% and 7.32% respectively in group and person recognition accuracy on CAD2, as well as an improvement of 6.63 in MOTP on HiEve. Pre-training with our synthetic data also leads to faster model convergence on downstream tasks (up to 6.8% faster). Moreover, M3Act opens new research problems for 3D group activity generation. We release M3Act3D, an 87.6-hour 3D motion dataset of human activities with larger group sizes and higher complexity of inter-person interactions than previous multi-person datasets. We define multiple metrics and propose a competitive baseline for the novel task.
All-intra rate control using low complexity video features for Versatile Video Coding
Authors: Vignesh V Menon, Anastasia Henkel, Prajit T Rajendran, Christian R. Helmrich, Adam Wieckowski, Benjamin Bross, Christian Timmerer, Detlev Marpe
Abstract
Versatile Video Coding (VVC) allows for large compression efficiency gains over its predecessor, High Efficiency Video Coding (HEVC). The added efficiency comes at the cost of increased runtime complexity, especially for encoding. It is thus highly relevant to explore all available runtime reduction options. This paper proposes a novel first pass for two-pass rate control in all-intra configuration, using low-complexity video analysis and a Random Forest (RF)-based machine learning model to derive the data required for driving the second pass. The proposed method is validated using VVenC, an open and optimized VVC encoder. Compared to the default two-pass rate control algorithm in VVenC, the proposed method achieves around 32% reduction in encoding time for the preset faster, while on average only causing 2% BD-rate increase and achieving similar rate control accuracy.
Sampling weights of deep neural networks
Authors: Erik Lien Bolager, Iryna Burak, Chinmay Datar, Qing Sun, Felix Dietrich
Abstract
We introduce a probability distribution, combined with an efficient sampling algorithm, for weights and biases of fully-connected neural networks. In a supervised learning context, no iterative optimization or gradient computations of internal network parameters are needed to obtain a trained network. The sampling is based on the idea of random feature models. However, instead of a data-agnostic distribution, e.g., a normal distribution, we use both the input and the output training data of the supervised learning problem to sample both shallow and deep networks. We prove that the sampled networks we construct are universal approximators. We also show that our sampling scheme is invariant to rigid body transformations and scaling of the input data. This implies many popular pre-processing techniques are no longer required. For Barron functions, we show that the $L^2$-approximation error of sampled shallow networks decreases with the square root of the number of neurons. In numerical experiments, we demonstrate that sampled networks achieve comparable accuracy as iteratively trained ones, but can be constructed orders of magnitude faster. Our test cases involve a classification benchmark from OpenML, sampling of neural operators to represent maps in function spaces, and transfer learning using well-known architectures.
A Formal Perspective on Byte-Pair Encoding
Authors: Vilém Zouhar, Clara Meister, Juan Luis Gastaldi, Li Du, Tim Vieira, Mrinmaya Sachan, Ryan Cotterell
Subjects: Computation and Language (cs.CL); Optimization and Control (math.OC)
Abstract
Byte-Pair Encoding (BPE) is a popular algorithm used for tokenizing data in NLP, despite being devised initially as a compression method. BPE appears to be a greedy algorithm at face value, but the underlying optimization problem that BPE seeks to solve has not yet been laid down. We formalize BPE as a combinatorial optimization problem. Via submodular functions, we prove that the iterative greedy version is a $\frac{1}{{\sigma(\boldsymbol{\mu}^\star)}}(1-e^{-{\sigma(\boldsymbol{\mu}^\star)}})$-approximation of an optimal merge sequence, where ${\sigma(\boldsymbol{\mu}^\star)}$ is the total backward curvature with respect to the optimal merge sequence $\boldsymbol{\mu}^\star$. Empirically the lower bound of the approximation is $\approx 0.37$. We provide a faster implementation of BPE which improves the runtime complexity from $\mathcal{O}\left(N M\right)$ to $\mathcal{O}\left(N \log M\right)$, where $N$ is the sequence length and $M$ is the merge count. Finally, we optimize the brute-force algorithm for optimal BPE using memoization.
ICDaeLST: Intensity-Controllable Detail Attention-enhanced for Lightweight Fast Style Transfer
Authors: Jiang Shi Qi
Subjects: Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV)
Abstract
The mainstream style transfer methods usually use pre-trained deep convolutional neural network (VGG) models as encoders, or use more complex model structures to achieve better style transfer effects. This leads to extremely slow processing speeds for practical tasks due to limited resources or higher resolution image processing, such as 4K images, severely hindering the practical application value of style transfer models. We introduce a lightweight and fast styletransfer model with controllable detail attention enhancement, named ICDaeLST. The model adopts a minimal, shallow, and small architecture, forming a very compact lightweight model for efficient forward inference. Although its structure is simple and has limited parameters, we achieve better overall color and texture structure matching by introducing a style discriminator, design additional global semantic invariance loss to preserve the semantic and structural information of the content image from a high-level global perspective, and design a shallow detail attention enhancement module to preserve the detail information of the content image from a low-level detail perspective. We also achieve controllable intensity during inference for the first time (adjusting the degree of detail retention and texture structure transfer based on subjective judgment) to meet different users' subjective evaluation of stylization effects. Compared with the current best-performing and most lightweight models, our model achieves better style transfer quality and better content structure and detail retention, while having a smaller model size (17-250 times smaller) and faster speed (0.26-6.5 times faster), and achieves the fastest processing speed of 0.38s on 4K high-resolution images.
SWAT: A System-Wide Approach to Tunable Leakage Mitigation in Encrypted Data Stores
Authors: Leqian Zheng, Lei Xu, Cong Wang, Sheng Wang, Yuke Hu, Zhan Qin, Feifei Li, Kui Ren
Abstract
Numerous studies have underscored the significant privacy risks associated with various leakage patterns in encrypted data stores. Most existing systems that conceal leakage either (1) incur substantial overheads, (2) focus on specific subsets of leakage patterns, or (3) apply the same security notion across various workloads, thereby impeding the attainment of fine-tuned privacy-efficiency trade-offs. In light of various detrimental leakage patterns, this paper starts with an investigation into which specific leakage patterns require our focus respectively in the contexts of key-value, range-query, and dynamic workloads. Subsequently, we introduce new security notions tailored to the specific privacy requirements of these workloads. Accordingly, we present, SWAT, an efficient construction that progressively enables these workloads, while provably mitigating system-wide leakage via a suite of algorithms with tunable privacy-efficiency trade-offs. We conducted extensive experiments and compiled a detailed result analysis, showing the efficiency of our solution. SWAT is about $10.6\times$ slower than an encryption-only data store that reveals various leakage patterns and is $31.6\times$ faster than a trivially zero-leakage solution. Meanwhile, the performance of SWAT remains highly competitive compared to other designs that mitigate specific types of leakage.
FogROS2-SGC: A ROS2 Cloud Robotics Platform for Secure Global Connectivity
Authors: Kaiyuan Chen, Ryan Hoque, Karthik Dharmarajan, Edith LLontop, Simeon Adebola, Jeffrey Ichnowski, John Kubiatowicz, Ken Goldberg
Abstract
The Robot Operating System (ROS2) is the most widely used software platform for building robotics applications. FogROS2 extends ROS2 to allow robots to access cloud computing on demand. However, ROS2 and FogROS2 assume that all robots are locally connected and that each robot has full access and control of the other robots. With applications like distributed multi-robot systems, remote robot control, and mobile robots, robotics increasingly involves the global Internet and complex trust management. Existing approaches for connecting disjoint ROS2 networks lack key features such as security, compatibility, efficiency, and ease of use. We introduce FogROS2-SGC, an extension of FogROS2 that can effectively connect robot systems across different physical locations, networks, and Data Distribution Services (DDS). With globally unique and location-independent identifiers, FogROS2-SGC securely and efficiently routes data between robotics components around the globe. FogROS2-SGC is agnostic to the ROS2 distribution and configuration, is compatible with non-ROS2 software, and seamlessly extends existing ROS2 applications without any code modification. Experiments suggest FogROS2-SGC is 19x faster than rosbridge (a ROS2 package with comparable features, but lacking security). We also apply FogROS2-SGC to 4 robots and compute nodes that are 3600km apart. Videos and code are available on the project website https://sites.google.com/view/fogros2-sgc.
Keyword: mobile
DNA-TEQ: An Adaptive Exponential Quantization of Tensors for DNN Inference
Authors: Bahareh Khabbazan, Marc Riera, Antonio González
Abstract
Quantization is commonly used in Deep Neural Networks (DNNs) to reduce the storage and computational complexity by decreasing the arithmetical precision of activations and weights, a.k.a. tensors. Efficient hardware architectures employ linear quantization to enable the deployment of recent DNNs onto embedded systems and mobile devices. However, linear uniform quantization cannot usually reduce the numerical precision to less than 8 bits without sacrificing high performance in terms of model accuracy. The performance loss is due to the fact that tensors do not follow uniform distributions. In this paper, we show that a significant amount of tensors fit into an exponential distribution. Then, we propose DNA-TEQ to exponentially quantize DNN tensors with an adaptive scheme that achieves the best trade-off between numerical precision and accuracy loss. The experimental results show that DNA-TEQ provides a much lower quantization bit-width compared to previous proposals, resulting in an average compression ratio of 40% over the linear INT8 baseline, with negligible accuracy loss and without retraining the DNNs. Besides, DNA-TEQ leads the way in performing dot-product operations in the exponential domain, which saves 66% of energy consumption on average for a set of widely used DNNs.
Coordinating Multiple Resources for Optimal Postdisaster Operation of Interdependent Electric Power and Natural Gas Distribution Systems
Authors: Wei Wang, Kaigui Xie, Hongbin Wang, Xingzhe Hou, Tao Chen, Hongzhou Chen, Yufei He
Abstract
Electric power and natural gas systems are not separated but rather are increasingly connected physically and functionally interdependent due to the continuing development of natural gas-fired generation and gas industry electrification. Such interdependency makes these two systems interact with each other when responding to disasters. The aggravated risk of cascading failures across the two systems has been exposed in recent energy crises, highlighting the significance of preparing these interdependent systems against disasters and helping their impacted services quickly recover. This promotes us to treat power and gas systems as one whole to fully capture their interactive behaviors. In this paper, we focus on the interdependent electric power and natural gas distribution systems (IENDS) and propose a "supply - demand - repair" strategy to comprehensively help the IENDS tide over the emergency periods after disasters by coordinating mobile or stationary emergency resources for various uses. Specifically, 1) on the supply side, the fuel supply issue of different types of generators for emergency use is considered and the fuel delivery process among different fuel facilities is mathematically formulated; 2) on the demand side, a zonewise method is proposed for integrated dispatch of power and gas demand responses; and 3) in the repair process, a varying efficiency related to the repair units at work is introduced to accurately model repairs. The proposed strategy is formulated into a mixed-integer second-order cone programming model to obtain a globally optimal decision of deploying all of those resources in a coordinated and organized manner. A series of case studies based on test systems are conducted to validate the effectiveness of the proposed strategy.
On the Impact of Interruptions During Multi-Robot Supervision Tasks
Authors: Abhinav Dahiya, Yifan Cai, Oliver Schneider, Stephen L. Smith
Abstract
Human supervisors in multi-robot systems are primarily responsible for monitoring robots, but can also be assigned with secondary tasks. These tasks can act as interruptions and can be categorized as either intrinsic, i.e., being directly related to the monitoring task, or extrinsic, i.e., being unrelated. In this paper, we investigate the impact of these two types of interruptions through a user study ($N=39$), where participants monitor a number of remote mobile robots while intermittently being interrupted by either a robot fault correction task (intrinsic) or a messaging task (extrinsic). We find that task performance of participants does not change significantly with the interruptions but depends greatly on the number of robots. However, interruptions result in an increase in perceived workload, and extrinsic interruptions have a more negative effect on workload across all NASA-TLX scales. Participants also reported switching between extrinsic interruptions and the primary task to be more difficult compared to the intrinsic interruption case. Statistical significance of these results is confirmed using ANOVA and one-sample t-test. These findings suggest that when deciding task assignment in such supervision systems, one should limit interruptions from secondary tasks, especially extrinsic ones, in order to limit user workload.
Abstract
Perception algorithms that provide estimates of their uncertainty are crucial to the development of autonomous robots that can operate in challenging and uncontrolled environments. Such perception algorithms provide the means for having risk-aware robots that reason about the probability of successfully completing a task when planning. There exist perception algorithms that come with models of their uncertainty; however, these models are often developed with assumptions, such as perfect data associations, that do not hold in the real world. Hence the resultant estimated uncertainty is a weak lower bound. To tackle this problem we present introspective perception - a novel approach for predicting accurate estimates of the uncertainty of perception algorithms deployed on mobile robots. By exploiting sensing redundancy and consistency constraints naturally present in the data collected by a mobile robot, introspective perception learns an empirical model of the error distribution of perception algorithms in the deployment environment and in an autonomously supervised manner. In this paper, we present the general theory of introspective perception and demonstrate successful implementations for two different perception tasks. We provide empirical results on challenging real-robot data for introspective stereo depth estimation and introspective visual simultaneous localization and mapping and show that they learn to predict their uncertainty with high accuracy and leverage this information to significantly reduce state estimation errors for an autonomous mobile robot.
TacMMs: Tactile Mobile Manipulators for Warehouse Automation
Authors: Zhuochao He, Xuyang Zhang, Simon Jones, Sabine Hauert, Dandan Zhang, Nathan F. Lepora
Abstract
Multi-robot platforms are playing an increasingly important role in warehouse automation for efficient goods transport. This paper proposes a novel customization of a multi-robot system, called Tactile Mobile Manipulators (TacMMs). Each TacMM integrates a soft optical tactile sensor and a mobile robot with a load-lifting mechanism, enabling cooperative transportation in tasks requiring coordinated physical interaction. More specifically, we mount the TacTip (biomimetic optical tactile sensor) on the Distributed Organisation and Transport System (DOTS) mobile robot. The tactile information then helps the mobile robots adjust the relative robot-object pose, thereby increasing the efficiency of load-lifting tasks. This study compares the performance of using two TacMMs with tactile perception with traditional vision-based pose adjustment for load-lifting. The results show that the average success rate of the TacMMs (66%) is improved over a purely visual-based method (34%), with a larger improvement when the mass of the load was non-uniformly distributed. Although this initial study considers two TacMMs, we expect the benefits of tactile perception to extend to multiple mobile robots. Website: https://sites.google.com/view/tacmms
A Framework for Securing Health Information Using Blockchain in Cloud Hosted Cyber Physical Systems
Abstract
Electronic Health Records (EHRs) have undergone numerous technical improvements in recent years, including the incorporation of mobile devices with the cloud computing technologies to facilitate medical data exchanges between patients and the healthcare professionals. This cutting-edge architecture enables cyber physical systems housed in the cloud to provide healthcare services with minimal operational costs, high flexibility, security, and EHR accessibility. If patient health information is stored in the hospital database, there will always be a risk of intrusion, i.e., unauthorized file access and information modification by attackers. To address this concern, we propose a decentralized EHR system based on Blockchain technology. To facilitate secure EHR exchange across various patients and medical providers, we develop a reliable access control method based on smart contracts. We incorporate Cryptocurrency, specifically Ethereum, in the suggested system to protect sensitive health information from potential attackers. In our suggested approach, both physicians and patients are required to be authenticated. Patients can register, and a block with a unique hash value will be generated. Once the patient discusses the disease with the physician, the physician can check the patient's condition and offer drugs. For experimental findings, we employ the public Block chain Ganache and solidity remix-based smart contracts to protect privacy. Ethers are used as the crypto currencies.
FogROS2-SGC: A ROS2 Cloud Robotics Platform for Secure Global Connectivity
Authors: Kaiyuan Chen, Ryan Hoque, Karthik Dharmarajan, Edith LLontop, Simeon Adebola, Jeffrey Ichnowski, John Kubiatowicz, Ken Goldberg
Abstract
The Robot Operating System (ROS2) is the most widely used software platform for building robotics applications. FogROS2 extends ROS2 to allow robots to access cloud computing on demand. However, ROS2 and FogROS2 assume that all robots are locally connected and that each robot has full access and control of the other robots. With applications like distributed multi-robot systems, remote robot control, and mobile robots, robotics increasingly involves the global Internet and complex trust management. Existing approaches for connecting disjoint ROS2 networks lack key features such as security, compatibility, efficiency, and ease of use. We introduce FogROS2-SGC, an extension of FogROS2 that can effectively connect robot systems across different physical locations, networks, and Data Distribution Services (DDS). With globally unique and location-independent identifiers, FogROS2-SGC securely and efficiently routes data between robotics components around the globe. FogROS2-SGC is agnostic to the ROS2 distribution and configuration, is compatible with non-ROS2 software, and seamlessly extends existing ROS2 applications without any code modification. Experiments suggest FogROS2-SGC is 19x faster than rosbridge (a ROS2 package with comparable features, but lacking security). We also apply FogROS2-SGC to 4 robots and compute nodes that are 3600km apart. Videos and code are available on the project website https://sites.google.com/view/fogros2-sgc.
Keyword: pruning
An Efficient Sparse Inference Software Accelerator for Transformer-based Language Models on CPUs
Authors: Haihao Shen, Hengyu Meng, Bo Dong, Zhe Wang, Ofir Zafrir, Yi Ding, Yu Luo, Hanwen Chang, Qun Gao, Ziheng Wang, Guy Boudoukh, Moshe Wasserblat
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Abstract
In recent years, Transformer-based language models have become the standard approach for natural language processing tasks. However, stringent throughput and latency requirements in industrial applications are limiting their adoption. To mitigate the gap, model compression techniques such as structured pruning are being used to improve inference efficiency. However, most existing neural network inference runtimes lack adequate support for structured sparsity. In this paper, we propose an efficient sparse deep learning inference software stack for Transformer-based language models where the weights are pruned with constant block size. Our sparse software accelerator leverages Intel Deep Learning Boost to maximize the performance of sparse matrix - dense matrix multiplication (commonly abbreviated as SpMM) on CPUs. Our SpMM kernel outperforms the existing sparse libraries (oneMKL, TVM, and LIBXSMM) by an order of magnitude on a wide range of GEMM shapes under 5 representative sparsity ratios (70%, 75%, 80%, 85%, 90%). Moreover, our SpMM kernel shows up to 5x speedup over dense GEMM kernel of oneDNN, a well-optimized dense library widely used in industry. We apply our sparse accelerator on widely-used Transformer-based language models including Bert-Mini, DistilBERT, Bert-Base, and BERT-Large. Our sparse inference software shows up to 1.5x speedup over Neural Magic's Deepsparse under same configurations on Xeon on Amazon Web Services under proxy production latency constraints. We also compare our solution with two framework-based inference solutions, ONNX Runtime and PyTorch, and demonstrate up to 37x speedup over ONNX Runtime and 345x over PyTorch on Xeon under the latency constraints. All the source code is publicly available on Github: https://github.com/intel/intel-extension-for-transformers.
Abstract
We introduce NaturalInversion, a novel model inversion-based method to synthesize images that agrees well with the original data distribution without using real data. In NaturalInversion, we propose: (1) a Feature Transfer Pyramid which uses enhanced image prior of the original data by combining the multi-scale feature maps extracted from the pre-trained classifier, (2) a one-to-one approach generative model where only one batch of images are synthesized by one generator to bring the non-linearity to optimization and to ease the overall optimizing process, (3) learnable Adaptive Channel Scaling parameters which are end-to-end trained to scale the output image channel to utilize the original image prior further. With our NaturalInversion, we synthesize images from classifiers trained on CIFAR-10/100 and show that our images are more consistent with original data distribution than prior works by visualization and additional analysis. Furthermore, our synthesized images outperform prior works on various applications such as knowledge distillation and pruning, demonstrating the effectiveness of our proposed method.
Rapid-INR: Storage Efficient CPU-free DNN Training Using Implicit Neural Representation
Authors: Hanqiu Chen, Hang Yang, Stephen BR Fitzmeyer, Cong Hao
Abstract
Implicit Neural Representation (INR) is an innovative approach for representing complex shapes or objects without explicitly defining their geometry or surface structure. Instead, INR represents objects as continuous functions. Previous research has demonstrated the effectiveness of using neural networks as INR for image compression, showcasing comparable performance to traditional methods such as JPEG. However, INR holds potential for various applications beyond image compression. This paper introduces Rapid-INR, a novel approach that utilizes INR for encoding and compressing images, thereby accelerating neural network training in computer vision tasks. Our methodology involves storing the whole dataset directly in INR format on a GPU, mitigating the significant data communication overhead between the CPU and GPU during training. Additionally, the decoding process from INR to RGB format is highly parallelized and executed on-the-fly. To further enhance compression, we propose iterative and dynamic pruning, as well as layer-wise quantization, building upon previous work. We evaluate our framework on the image classification task, utilizing the ResNet-18 backbone network and three commonly used datasets with varying image sizes. Rapid-INR reduces memory consumption to only 5% of the original dataset size and achieves a maximum 6$\times$ speedup over the PyTorch training pipeline, as well as a maximum 1.2x speedup over the DALI training pipeline, with only a marginal decrease in accuracy. Importantly, Rapid-INR can be readily applied to other computer vision tasks and backbone networks with reasonable engineering efforts. Our implementation code is publicly available at https://anonymous.4open.science/r/INR-4BF7.
Sparse Model Soups: A Recipe for Improved Pruning via Model Averaging
Authors: Max Zimmer, Christoph Spiegel, Sebastian Pokutta
Abstract
Neural networks can be significantly compressed by pruning, leading to sparse models requiring considerably less storage and floating-point operations while maintaining predictive performance. Model soups (Wortsman et al., 2022) improve generalization and out-of-distribution performance by averaging the parameters of multiple models into a single one without increased inference time. However, identifying models in the same loss basin to leverage both sparsity and parameter averaging is challenging, as averaging arbitrary sparse models reduces the overall sparsity due to differing sparse connectivities. In this work, we address these challenges by demonstrating that exploring a single retraining phase of Iterative Magnitude Pruning (IMP) with varying hyperparameter configurations, such as batch ordering or weight decay, produces models that are suitable for averaging and share the same sparse connectivity by design. Averaging these models significantly enhances generalization performance compared to their individual components. Building on this idea, we introduce Sparse Model Soups (SMS), a novel method for merging sparse models by initiating each prune-retrain cycle with the averaged model of the previous phase. SMS maintains sparsity, exploits sparse network benefits being modular and fully parallelizable, and substantially improves IMP's performance. Additionally, we demonstrate that SMS can be adapted to enhance the performance of state-of-the-art pruning during training approaches.
Can Machines Garden? Systematically Comparing the AlphaGarden vs. Professional Horticulturalists
Authors: Simeon Adebola, Rishi Parikh, Mark Presten, Satvik Sharma, Shrey Aeron, Ananth Rao, Sandeep Mukherjee, Tomson Qu, Christina Wistrom, Eugen Solowjow, Ken Goldberg
Abstract
The AlphaGarden is an automated testbed for indoor polyculture farming which combines a first-order plant simulator, a gantry robot, a seed planting algorithm, plant phenotyping and tracking algorithms, irrigation sensors and algorithms, and custom pruning tools and algorithms. In this paper, we systematically compare the performance of the AlphaGarden to professional horticulturalists on the staff of the UC Berkeley Oxford Tract Greenhouse. The humans and the machine tend side-by-side polyculture gardens with the same seed arrangement. We compare performance in terms of canopy coverage, plant diversity, and water consumption. Results from two 60-day cycles suggest that the automated AlphaGarden performs comparably to professional horticulturalists in terms of coverage and diversity, and reduces water consumption by as much as 44%. Code, videos, and datasets are available at https://sites.google.com/berkeley.edu/systematiccomparison.
Keyword: diffusion
Solver algorithm for stabilized space-time formulation of advection-dominated diffusion problem
Authors: Marcin Łoś, Paulina Sepulveda-Salas, Maciej Paszyński
Abstract
This article shows how to develop an efficient solver for a stabilized numerical space-time formulation of the advection-dominated diffusion transient equation. At the discrete space-time level, we approximate the solution by using higher-order continuous B-spline basis functions in its spatial and temporal dimensions. This problem is very difficult to solve numerically using the standard Galerkin finite element method due to artificial oscillations present when the advection term dominates the diffusion term. However, a first-order constraint least-square formulation allows us to obtain numerical solutions avoiding oscillations. The advantages of space-time formulations are the use of high-order methods and the feasibility of developing space-time mesh adaptive techniques on well-defined discrete problems. We develop a solver for a least-square formulation to obtain a stabilized and symmetric problem on finite element meshes. The computational cost of our solver is bounded by the cost of the inversion of the space-time mass and stiffness (with one value fixed at a point) matrices and the cost of the GMRES solver applied for the symmetric and positive definite problem. We illustrate our findings on an advection-dominated diffusion space-time model problem and present two numerical examples: one with isogeometric analysis discretizations and the second one with an adaptive space-time finite element method.
DiffusionSTR: Diffusion Model for Scene Text Recognition
Authors: Masato Fujitake
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
This paper presents Diffusion Model for Scene Text Recognition (DiffusionSTR), an end-to-end text recognition framework using diffusion models for recognizing text in the wild. While existing studies have viewed the scene text recognition task as an image-to-text transformation, we rethought it as a text-text one under images in a diffusion model. We show for the first time that the diffusion model can be applied to text recognition. Furthermore, experimental results on publicly available datasets show that the proposed method achieves competitive accuracy compared to state-of-the-art methods.
New twofold saddle-point formulations for Biot poroelasticity with porosity-dependent permeability
Authors: Bishnu P. Lamichhane, Ricardo Ruiz-Baier, Segundo Villa-Fuentes
Abstract
We propose four-field and five-field Hu--Washizu-type mixed formulations for nonlinear poroelasticity -- a coupled fluid diffusion and solid deformation process -- considering that the permeability depends on a linear combination between fluid pressure and dilation. As the determination of the physical strains is necessary, the first formulation is written in terms of the primal unknowns of solid displacement and pore fluid pressure as well as the poroelastic stress and the infinitesimal strain, and it considers strongly symmetric Cauchy stresses. The second formulation imposes stress symmetry in a weak sense and it requires the additional unknown of solid rotation tensor. We study the unique solvability of the problem using the Banach fixed-point theory, properties of twofold saddle-point problems, and the Banach--Ne\v{c}as--Babu\v{s}ka theory. We propose monolithic Galerkin discretisations based on conforming Arnold--Winther for poroelastic stress and displacement, and either PEERS or Arnold--Falk--Winther finite element families for the stress-displacement-rotation field variables. The wellposedness of the discrete problem is established as well, and we show a priori error estimates in the natural norms. Some numerical examples are provided to confirm the rates of convergence predicted by the theory, and we also illustrate the use of the formulation in some typical tests in Biot poroelasticity.
Efficient Sobolev approximation of linear parabolic PDEs in high dimensions
Abstract
In this paper, we study the error in first order Sobolev norm in the approximation of solutions to linear parabolic PDEs. We use a Monte Carlo Euler scheme obtained from combining the Feynman--Kac representation with a Euler discretization of the underlying stochastic process. We derive approximation rates depending on the time-discretization, the number of Monte Carlo simulations, and the dimension. In particular, we show that the Monte Carlo Euler scheme breaks the curse of dimensionality with respect to the first order Sobolev norm. Our argument is based on new estimates on the weak error of the Euler approximation of a diffusion process together with its derivative with respect to the initial condition. As a consequence, we obtain that neural networks are able to approximate solutions of linear parabolic PDEs in first order Sobolev norm without the curse of dimensionality if the coefficients of the PDEs admit an efficient approximation with neural networks.
SaGess: Sampling Graph Denoising Diffusion Model for Scalable Graph Generation
Authors: Stratis Limnios, Praveen Selvaraj, Mihai Cucuringu, Carsten Maple, Gesine Reinert, Andrew Elliott
Abstract
Over recent years, denoising diffusion generative models have come to be considered as state-of-the-art methods for synthetic data generation, especially in the case of generating images. These approaches have also proved successful in other applications such as tabular and graph data generation. However, due to computational complexity, to this date, the application of these techniques to graph data has been restricted to small graphs, such as those used in molecular modeling. In this paper, we propose SaGess, a discrete denoising diffusion approach, which is able to generate large real-world networks by augmenting a diffusion model (DiGress) with a generalized divide-and-conquer framework. The algorithm is capable of generating larger graphs by sampling a covering of subgraphs of the initial graph in order to train DiGress. SaGess then constructs a synthetic graph using the subgraphs that have been generated by DiGress. We evaluate the quality of the synthetic data sets against several competitor methods by comparing graph statistics between the original and synthetic samples, as well as evaluating the utility of the synthetic data set produced by using it to train a task-driven model, namely link prediction. In our experiments, SaGess, outperforms most of the one-shot state-of-the-art graph generating methods by a significant factor, both on the graph metrics and on the link prediction task.
PFB-Diff: Progressive Feature Blending Diffusion for Text-driven Image Editing
Abstract
Diffusion models have showcased their remarkable capability to synthesize diverse and high-quality images, sparking interest in their application for real image editing. However, existing diffusion-based approaches for local image editing often suffer from undesired artifacts due to the pixel-level blending of the noised target images and diffusion latent variables, which lack the necessary semantics for maintaining image consistency. To address these issues, we propose PFB-Diff, a Progressive Feature Blending method for Diffusion-based image editing. Unlike previous methods, PFB-Diff seamlessly integrates text-guided generated content into the target image through multi-level feature blending. The rich semantics encoded in deep features and the progressive blending scheme from high to low levels ensure semantic coherence and high quality in edited images. Additionally, we introduce an attention masking mechanism in the cross-attention layers to confine the impact of specific words to desired regions, further improving the performance of background editing. PFB-Diff can effectively address various editing tasks, including object/background replacement and object attribute editing. Our method demonstrates its superior performance in terms of image fidelity, editing accuracy, efficiency, and faithfulness to the original image, without the need for fine-tuning or training.
Whole-Body Exploration with a Manipulator Using Heat Equation
Authors: Cem Bilaloglu, Tobias Löw, Sylvain Calinon
Abstract
This paper presents a whole-body robot control method for exploring and probing a given region of interest. The ergodic control formalism behind such an exploration behavior consists of matching the time-averaged statistics of a robot trajectory with the spatial statistics of the target distribution. Most existing ergodic control approaches assume the robots/sensors as individual point agents moving in space. We introduce an approach exploiting multiple kinematically constrained agents on the whole-body of a robotic manipulator, where a consensus among the agents is found for generating control actions. To do so, we exploit an existing ergodic control formulation called heat equation-driven area coverage (HEDAC), combining local and global exploration on a potential field resulting from heat diffusion. Our approach extends HEDAC to applications where robots have multiple sensors on the whole-body (such as tactile skin) and use all sensors to optimally explore the given region. We show that our approach increases the exploration performance in terms of ergodicity and scales well to real-world problems using agents distributed on multiple robot links. We compare our method with HEDAC in kinematic simulation and demonstrate the applicability of an online exploration task with a 7-axis Franka Emika robot.
One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimization
Abstract
Single image 3D reconstruction is an important but challenging task that requires extensive knowledge of our natural world. Many existing methods solve this problem by optimizing a neural radiance field under the guidance of 2D diffusion models but suffer from lengthy optimization time, 3D inconsistency results, and poor geometry. In this work, we propose a novel method that takes a single image of any object as input and generates a full 360-degree 3D textured mesh in a single feed-forward pass. Given a single image, we first use a view-conditioned 2D diffusion model, Zero123, to generate multi-view images for the input view, and then aim to lift them up to 3D space. Since traditional reconstruction methods struggle with inconsistent multi-view predictions, we build our 3D reconstruction module upon an SDF-based generalizable neural surface reconstruction method and propose several critical training strategies to enable the reconstruction of 360-degree meshes. Without costly optimizations, our method reconstructs 3D shapes in significantly less time than existing methods. Moreover, our method favors better geometry, generates more 3D consistent results, and adheres more closely to the input image. We evaluate our approach on both synthetic data and in-the-wild images and demonstrate its superiority in terms of both mesh quality and runtime. In addition, our approach can seamlessly support the text-to-3D task by integrating with off-the-shelf text-to-image diffusion models.
DreamDiffusion: Generating High-Quality Images from Brain EEG Signals
Abstract
This paper introduces DreamDiffusion, a novel method for generating high-quality images directly from brain electroencephalogram (EEG) signals, without the need to translate thoughts into text. DreamDiffusion leverages pre-trained text-to-image models and employs temporal masked signal modeling to pre-train the EEG encoder for effective and robust EEG representations. Additionally, the method further leverages the CLIP image encoder to provide extra supervision to better align EEG, text, and image embeddings with limited EEG-image pairs. Overall, the proposed method overcomes the challenges of using EEG signals for image generation, such as noise, limited information, and individual differences, and achieves promising results. Quantitative and qualitative results demonstrate the effectiveness of the proposed method as a significant step towards portable and low-cost ``thoughts-to-image'', with potential applications in neuroscience and computer vision.
Diffusion-Jump GNNs: Homophiliation via Learnable Metric Filters
Authors: Ahmed Begga, Francisco Escolano, Miguel Angel Lozano, Edwin R. Hancock
Abstract
High-order Graph Neural Networks (HO-GNNs) have been developed to infer consistent latent spaces in the heterophilic regime, where the label distribution is not correlated with the graph structure. However, most of the existing HO-GNNs are hop-based, i.e., they rely on the powers of the transition matrix. As a result, these architectures are not fully reactive to the classification loss and the achieved structural filters have static supports. In other words, neither the filters' supports nor their coefficients can be learned with these networks. They are confined, instead, to learn combinations of filters. To address the above concerns, we propose Diffusion-jump GNNs a method relying on asymptotic diffusion distances that operates on jumps. A diffusion-pump generates pairwise distances whose projections determine both the support and coefficients of each structural filter. These filters are called jumps because they explore a wide range of scales in order to find bonds between scattered nodes with the same label. Actually, the full process is controlled by the classification loss. Both the jumps and the diffusion distances react to classification errors (i.e. they are learnable). Homophiliation, i.e., the process of learning piecewise smooth latent spaces in the heterophilic regime, is formulated as a Dirichlet problem: the known labels determine the border nodes and the diffusion-pump ensures a minimal deviation of the semi-supervised grouping from a canonical unsupervised grouping. This triggers the update of both the diffusion distances and, consequently, the jumps in order to minimize the classification error. The Dirichlet formulation has several advantages. It leads to the definition of structural heterophily, a novel measure beyond edge heterophily. It also allows us to investigate links with (learnable) diffusion distances, absorbing random walks and stochastic diffusion.
Abstract
Spiking neural networks (SNNs) have ultra-low energy consumption and high biological plausibility due to their binary and bio-driven nature compared with artificial neural networks (ANNs). While previous research has primarily focused on enhancing the performance of SNNs in classification tasks, the generative potential of SNNs remains relatively unexplored. In our paper, we put forward Spiking Denoising Diffusion Probabilistic Models (SDDPM), a new class of SNN-based generative models that achieve high sample quality. To fully exploit the energy efficiency of SNNs, we propose a purely Spiking U-Net architecture, which achieves comparable performance to its ANN counterpart using only 4 time steps, resulting in significantly reduced energy consumption. Extensive experimental results reveal that our approach achieves state-of-the-art on the generative tasks and substantially outperforms other SNN-based generative models, achieving up to $12\times$ and $6\times$ improvement on the CIFAR-10 and the CelebA datasets, respectively. Moreover, we propose a threshold-guided strategy that can further improve the performances by 16.7% in a training-free manner. The SDDPM symbolizes a significant advancement in the field of SNN generation, injecting new perspectives and potential avenues of exploration.
Learning Structure-Guided Diffusion Model for 2D Human Pose Estimation
Abstract
One of the mainstream schemes for 2D human pose estimation (HPE) is learning keypoints heatmaps by a neural network. Existing methods typically improve the quality of heatmaps by customized architectures, such as high-resolution representation and vision Transformers. In this paper, we propose \textbf{DiffusionPose}, a new scheme that formulates 2D HPE as a keypoints heatmaps generation problem from noised heatmaps. During training, the keypoints are diffused to random distribution by adding noises and the diffusion model learns to recover ground-truth heatmaps from noised heatmaps with respect to conditions constructed by image feature. During inference, the diffusion model generates heatmaps from initialized heatmaps in a progressive denoising way. Moreover, we further explore improving the performance of DiffusionPose with conditions from human structural information. Extensive experiments show the prowess of our DiffusionPose, with improvements of 1.6, 1.2, and 1.2 mAP on widely-used COCO, CrowdPose, and AI Challenge datasets, respectively.
Michelangelo: Conditional 3D Shape Generation based on Shape-Image-Text Aligned Latent Representation
Authors: Zibo Zhao, Wen Liu, Xin Chen, Xianfang Zeng, Rui Wang, Pei Cheng, Bin Fu, Tao Chen, Gang Yu, Shenghua Gao
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
We present a novel alignment-before-generation approach to tackle the challenging task of generating general 3D shapes based on 2D images or texts. Directly learning a conditional generative model from images or texts to 3D shapes is prone to producing inconsistent results with the conditions because 3D shapes have an additional dimension whose distribution significantly differs from that of 2D images and texts. To bridge the domain gap among the three modalities and facilitate multi-modal-conditioned 3D shape generation, we explore representing 3D shapes in a shape-image-text-aligned space. Our framework comprises two models: a Shape-Image-Text-Aligned Variational Auto-Encoder (SITA-VAE) and a conditional Aligned Shape Latent Diffusion Model (ASLDM). The former model encodes the 3D shapes into the shape latent space aligned to the image and text and reconstructs the fine-grained 3D neural fields corresponding to given shape embeddings via the transformer-based decoder. The latter model learns a probabilistic mapping function from the image or text space to the latent shape space. Our extensive experiments demonstrate that our proposed approach can generate higher-quality and more diverse 3D shapes that better semantically conform to the visual or textural conditional inputs, validating the effectiveness of the shape-image-text-aligned space for cross-modality 3D shape generation.
ID-Pose: Sparse-view Camera Pose Estimation by Inverting Diffusion Models
Authors: Weihao Cheng, Yan-Pei Cao, Ying Shan
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Given sparse views of an object, estimating their camera poses is a long-standing and intractable problem. We harness the pre-trained diffusion model of novel views conditioned on viewpoints (Zero-1-to-3). We present ID-Pose which inverses the denoising diffusion process to estimate the relative pose given two input images. ID-Pose adds a noise on one image, and predicts the noise conditioned on the other image and a decision variable for the pose. The prediction error is used as the objective to find the optimal pose with the gradient descent method. ID-Pose can handle more than two images and estimate each of the poses with multiple image pairs from triangular relationships. ID-Pose requires no training and generalizes to real-world images. We conduct experiments using high-quality real-scanned 3D objects, where ID-Pose significantly outperforms state-of-the-art methods.
Filtered-Guided Diffusion: Fast Filter Guidance for Black-Box Diffusion Models
Authors: Zeqi Gu, Abe Davis
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent advances in diffusion-based generative models have shown incredible promise for Image-to-Image translation and editing. Most recent work in this space relies on additional training or architecture-specific adjustments to the diffusion process. In this work, we show that much of this low-level control can be achieved without additional training or any access to features of the diffusion model. Our method simply applies a filter to the input of each diffusion step based on the output of the previous step in an adaptive manner. Notably, this approach does not depend on any specific architecture or sampler and can be done without access to internal features of the network, making it easy to combine with other techniques, samplers, and diffusion architectures. Furthermore, it has negligible cost to performance, and allows for more continuous adjustment of guidance strength than other approaches. We show FGD offers a fast and strong baseline that is competitive with recent architecture-dependent approaches. Furthermore, FGD can also be used as a simple add-on to enhance the structural guidance of other state-of-the-art I2I methods. Finally, our derivation of this method helps to understand the impact of self attention, a key component of other recent architecture-specific I2I approaches, in a more architecture-independent way. Project page: https://github.com/jaclyngu/FilteredGuidedDiffusion
Generate Anything Anywhere in Any Scene
Authors: Yuheng Li, Haotian Liu, Yangming Wen, Yong Jae Lee
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Text-to-image diffusion models have attracted considerable interest due to their wide applicability across diverse fields. However, challenges persist in creating controllable models for personalized object generation. In this paper, we first identify the entanglement issues in existing personalized generative models, and then propose a straightforward and efficient data augmentation training strategy that guides the diffusion model to focus solely on object identity. By inserting the plug-and-play adapter layers from a pre-trained controllable diffusion model, our model obtains the ability to control the location and size of each generated personalized object. During inference, we propose a regionally-guided sampling technique to maintain the quality and fidelity of the generated images. Our method achieves comparable or superior fidelity for personalized objects, yielding a robust, versatile, and controllable text-to-image diffusion model that is capable of generating realistic and personalized images. Our approach demonstrates significant potential for various applications, such as those in art, entertainment, and advertising design.
Keyword: adaptive
DNA-TEQ: An Adaptive Exponential Quantization of Tensors for DNN Inference
Authors: Bahareh Khabbazan, Marc Riera, Antonio González
Abstract
Quantization is commonly used in Deep Neural Networks (DNNs) to reduce the storage and computational complexity by decreasing the arithmetical precision of activations and weights, a.k.a. tensors. Efficient hardware architectures employ linear quantization to enable the deployment of recent DNNs onto embedded systems and mobile devices. However, linear uniform quantization cannot usually reduce the numerical precision to less than 8 bits without sacrificing high performance in terms of model accuracy. The performance loss is due to the fact that tensors do not follow uniform distributions. In this paper, we show that a significant amount of tensors fit into an exponential distribution. Then, we propose DNA-TEQ to exponentially quantize DNN tensors with an adaptive scheme that achieves the best trade-off between numerical precision and accuracy loss. The experimental results show that DNA-TEQ provides a much lower quantization bit-width compared to previous proposals, resulting in an average compression ratio of 40% over the linear INT8 baseline, with negligible accuracy loss and without retraining the DNNs. Besides, DNA-TEQ leads the way in performing dot-product operations in the exponential domain, which saves 66% of energy consumption on average for a set of widely used DNNs.
Solver algorithm for stabilized space-time formulation of advection-dominated diffusion problem
Authors: Marcin Łoś, Paulina Sepulveda-Salas, Maciej Paszyński
Abstract
This article shows how to develop an efficient solver for a stabilized numerical space-time formulation of the advection-dominated diffusion transient equation. At the discrete space-time level, we approximate the solution by using higher-order continuous B-spline basis functions in its spatial and temporal dimensions. This problem is very difficult to solve numerically using the standard Galerkin finite element method due to artificial oscillations present when the advection term dominates the diffusion term. However, a first-order constraint least-square formulation allows us to obtain numerical solutions avoiding oscillations. The advantages of space-time formulations are the use of high-order methods and the feasibility of developing space-time mesh adaptive techniques on well-defined discrete problems. We develop a solver for a least-square formulation to obtain a stabilized and symmetric problem on finite element meshes. The computational cost of our solver is bounded by the cost of the inversion of the space-time mass and stiffness (with one value fixed at a point) matrices and the cost of the GMRES solver applied for the symmetric and positive definite problem. We illustrate our findings on an advection-dominated diffusion space-time model problem and present two numerical examples: one with isogeometric analysis discretizations and the second one with an adaptive space-time finite element method.
Deep Equilibrium Multimodal Fusion
Authors: Jinhong Ni, Yalong Bai, Wei Zhang, Ting Yao, Tao Mei
Subjects: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
Abstract
Multimodal fusion integrates the complementary information present in multiple modalities and has gained much attention recently. Most existing fusion approaches either learn a fixed fusion strategy during training and inference, or are only capable of fusing the information to a certain extent. Such solutions may fail to fully capture the dynamics of interactions across modalities especially when there are complex intra- and inter-modality correlations to be considered for informative multimodal fusion. In this paper, we propose a novel deep equilibrium (DEQ) method towards multimodal fusion via seeking a fixed point of the dynamic multimodal fusion process and modeling the feature correlations in an adaptive and recursive manner. This new way encodes the rich information within and across modalities thoroughly from low level to high level for efficacious downstream multimodal learning and is readily pluggable to various multimodal frameworks. Extensive experiments on BRCA, MM-IMDB, CMU-MOSI, SUN RGB-D, and VQA-v2 demonstrate the superiority of our DEQ fusion. More remarkably, DEQ fusion consistently achieves state-of-the-art performance on multiple multimodal benchmarks. The code will be released.
Abstract
We introduce NaturalInversion, a novel model inversion-based method to synthesize images that agrees well with the original data distribution without using real data. In NaturalInversion, we propose: (1) a Feature Transfer Pyramid which uses enhanced image prior of the original data by combining the multi-scale feature maps extracted from the pre-trained classifier, (2) a one-to-one approach generative model where only one batch of images are synthesized by one generator to bring the non-linearity to optimization and to ease the overall optimizing process, (3) learnable Adaptive Channel Scaling parameters which are end-to-end trained to scale the output image channel to utilize the original image prior further. With our NaturalInversion, we synthesize images from classifiers trained on CIFAR-10/100 and show that our images are more consistent with original data distribution than prior works by visualization and additional analysis. Furthermore, our synthesized images outperform prior works on various applications such as knowledge distillation and pruning, demonstrating the effectiveness of our proposed method.
Dynamic-Resolution Model Learning for Object Pile Manipulation
Abstract
Dynamics models learned from visual observations have shown to be effective in various robotic manipulation tasks. One of the key questions for learning such dynamics models is what scene representation to use. Prior works typically assume representation at a fixed dimension or resolution, which may be inefficient for simple tasks and ineffective for more complicated tasks. In this work, we investigate how to learn dynamic and adaptive representations at different levels of abstraction to achieve the optimal trade-off between efficiency and effectiveness. Specifically, we construct dynamic-resolution particle representations of the environment and learn a unified dynamics model using graph neural networks (GNNs) that allows continuous selection of the abstraction level. During test time, the agent can adaptively determine the optimal resolution at each model-predictive control (MPC) step. We evaluate our method in object pile manipulation, a task we commonly encounter in cooking, agriculture, manufacturing, and pharmaceutical applications. Through comprehensive evaluations both in the simulation and the real world, we show that our method achieves significantly better performance than state-of-the-art fixed-resolution baselines at the gathering, sorting, and redistribution of granular object piles made with various instances like coffee beans, almonds, corn, etc.
Metric-aligned Sample Selection and Critical Feature Sampling for Oriented Object Detection
Abstract
Arbitrary-oriented object detection is a relatively emerging but challenging task. Although remarkable progress has been made, there still remain many unsolved issues due to the large diversity of patterns in orientation, scale, aspect ratio, and visual appearance of objects in aerial images. Most of the existing methods adopt a coarse-grained fixed label assignment strategy and suffer from the inconsistency between the classification score and localization accuracy. First, to align the metric inconsistency between sample selection and regression loss calculation caused by fixed IoU strategy, we introduce affine transformation to evaluate the quality of samples and propose a distance-based label assignment strategy. The proposed metric-aligned selection (MAS) strategy can dynamically select samples according to the shape and rotation characteristic of objects. Second, to further address the inconsistency between classification and localization, we propose a critical feature sampling (CFS) module, which performs localization refinement on the sampling location for classification task to extract critical features accurately. Third, we present a scale-controlled smooth $L_1$ loss (SC-Loss) to adaptively select high quality samples by changing the form of regression loss function based on the statistics of proposals during training. Extensive experiments are conducted on four challenging rotated object detection datasets DOTA, FAIR1M-1.0, HRSC2016, and UCAS-AOD. The results show the state-of-the-art accuracy of the proposed detector.
Multi-Scenario Ranking with Adaptive Feature Learning
Authors: Yu Tian, Bofang Li, Si Chen, Xubin Li, Hongbo Deng, Jian Xu, Bo Zheng, Qian Wang, Chenliang Li
Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)
Abstract
Recently, Multi-Scenario Learning (MSL) is widely used in recommendation and retrieval systems in the industry because it facilitates transfer learning from different scenarios, mitigating data sparsity and reducing maintenance cost. These efforts produce different MSL paradigms by searching more optimal network structure, such as Auxiliary Network, Expert Network, and Multi-Tower Network. It is intuitive that different scenarios could hold their specific characteristics, activating the user's intents quite differently. In other words, different kinds of auxiliary features would bear varying importance under different scenarios. With more discriminative feature representations refined in a scenario-aware manner, better ranking performance could be easily obtained without expensive search for the optimal network structure. Unfortunately, this simple idea is mainly overlooked but much desired in real-world systems.Further analysis also validates the rationality of adaptive feature learning under a multi-scenario scheme. Moreover, our A/B test results on the Alibaba search advertising platform also demonstrate that Maria is superior in production environments.
Influence Maximization based on Simplicial Contagion Model in Hypergraph
Authors: Renquan Zhang, Ting Wei, Sen Pei
Subjects: Social and Information Networks (cs.SI); Physics and Society (physics.soc-ph)
Abstract
In recent years, the issue of node centrality has been actively and extensively explored due to its applications in product recommendations, opinion propagation, disease spread, and other areas involving maximizing node influence. This paper focuses on the problem of influence maximization on the Simplicial Contagion Model, using the susceptible-infectedrecovered (SIR) model as an example. To find practical solutions to this optimization problem, we have developed a theoretical framework based on message passing processes and conducted stability analysis of equilibrium solutions for the self-consistent equations. Furthermore, we introduce a metric called collective influence and propose an adaptive algorithm, known as the Collective Influence Adaptive (CIA), to identify influential propagators in the spreading process. This method has been validated on both synthetic hypergraphs and real hypergraphs, outperforming other competing heuristic methods.
RowPress: Amplifying Read Disturbance in Modern DRAM Chips
Authors: Haocong Luo, Ataberk Olgun, A. Giray Yağlıkçı, Yahya Can Tuğrul, Steve Rhyner, Meryem Banu Cavlak, Joël Lindegger, Mohammad Sadrosadati, Onur Mutlu
Subjects: Cryptography and Security (cs.CR); Hardware Architecture (cs.AR)
Abstract
Memory isolation is critical for system reliability, security, and safety. Unfortunately, read disturbance can break memory isolation in modern DRAM chips. For example, RowHammer is a well-studied read-disturb phenomenon where repeatedly opening and closing (i.e., hammering) a DRAM row many times causes bitflips in physically nearby rows. This paper experimentally demonstrates and analyzes another widespread read-disturb phenomenon, RowPress, in real DDR4 DRAM chips. RowPress breaks memory isolation by keeping a DRAM row open for a long period of time, which disturbs physically nearby rows enough to cause bitflips. We show that RowPress amplifies DRAM's vulnerability to read-disturb attacks by significantly reducing the number of row activations needed to induce a bitflip by one to two orders of magnitude under realistic conditions. In extreme cases, RowPress induces bitflips in a DRAM row when an adjacent row is activated only once. Our detailed characterization of 164 real DDR4 DRAM chips shows that RowPress 1) affects chips from all three major DRAM manufacturers, 2) gets worse as DRAM technology scales down to smaller node sizes, and 3) affects a different set of DRAM cells from RowHammer and behaves differently from RowHammer as temperature and access pattern changes. We demonstrate in a real DDR4-based system with RowHammer protection that 1) a user-level program induces bitflips by leveraging RowPress while conventional RowHammer cannot do so, and 2) a memory controller that adaptively keeps the DRAM row open for a longer period of time based on access pattern can facilitate RowPress-based attacks. To prevent bitflips due to RowPress, we describe and evaluate a new methodology that adapts existing RowHammer mitigation techniques to also mitigate RowPress with low additional performance overhead. We open source all our code and data to facilitate future research on RowPress.
Re-Rank - Expand - Repeat: Adaptive Query Expansion for Document Retrieval Using Words and Entities
Authors: Iain Mackie, Shubham Chatterjee, Sean MacAvaney, Jeffrey Dalton
Abstract
Sparse and dense pseudo-relevance feedback (PRF) approaches perform poorly on challenging queries due to low precision in first-pass retrieval. However, recent advances in neural language models (NLMs) can re-rank relevant documents to top ranks, even when few are in the re-ranking pool. This paper first addresses the problem of poor pseudo-relevance feedback by simply applying re-ranking prior to query expansion and re-executing this query. We find that this change alone can improve the retrieval effectiveness of sparse and dense PRF approaches by 5-8%. Going further, we propose a new expansion model, Latent Entity Expansion (LEE), a fine-grained word and entity-based relevance modelling incorporating localized features. Finally, we include an "adaptive" component to the retrieval process, which iteratively refines the re-ranking pool during scoring using the expansion model, i.e. we "re-rank - expand - repeat". Using LEE, we achieve (to our knowledge) the best NDCG, MAP and R@1000 results on the TREC Robust 2004 and CODEC adhoc document datasets, demonstrating a significant advancement in expansion effectiveness.
Filtered-Guided Diffusion: Fast Filter Guidance for Black-Box Diffusion Models
Authors: Zeqi Gu, Abe Davis
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent advances in diffusion-based generative models have shown incredible promise for Image-to-Image translation and editing. Most recent work in this space relies on additional training or architecture-specific adjustments to the diffusion process. In this work, we show that much of this low-level control can be achieved without additional training or any access to features of the diffusion model. Our method simply applies a filter to the input of each diffusion step based on the output of the previous step in an adaptive manner. Notably, this approach does not depend on any specific architecture or sampler and can be done without access to internal features of the network, making it easy to combine with other techniques, samplers, and diffusion architectures. Furthermore, it has negligible cost to performance, and allows for more continuous adjustment of guidance strength than other approaches. We show FGD offers a fast and strong baseline that is competitive with recent architecture-dependent approaches. Furthermore, FGD can also be used as a simple add-on to enhance the structural guidance of other state-of-the-art I2I methods. Finally, our derivation of this method helps to understand the impact of self attention, a key component of other recent architecture-specific I2I approaches, in a more architecture-independent way. Project page: https://github.com/jaclyngu/FilteredGuidedDiffusion
Keyword: quantization
DNA-TEQ: An Adaptive Exponential Quantization of Tensors for DNN Inference
Authors: Bahareh Khabbazan, Marc Riera, Antonio González
Abstract
Quantization is commonly used in Deep Neural Networks (DNNs) to reduce the storage and computational complexity by decreasing the arithmetical precision of activations and weights, a.k.a. tensors. Efficient hardware architectures employ linear quantization to enable the deployment of recent DNNs onto embedded systems and mobile devices. However, linear uniform quantization cannot usually reduce the numerical precision to less than 8 bits without sacrificing high performance in terms of model accuracy. The performance loss is due to the fact that tensors do not follow uniform distributions. In this paper, we show that a significant amount of tensors fit into an exponential distribution. Then, we propose DNA-TEQ to exponentially quantize DNN tensors with an adaptive scheme that achieves the best trade-off between numerical precision and accuracy loss. The experimental results show that DNA-TEQ provides a much lower quantization bit-width compared to previous proposals, resulting in an average compression ratio of 40% over the linear INT8 baseline, with negligible accuracy loss and without retraining the DNNs. Besides, DNA-TEQ leads the way in performing dot-product operations in the exponential domain, which saves 66% of energy consumption on average for a set of widely used DNNs.
Rapid-INR: Storage Efficient CPU-free DNN Training Using Implicit Neural Representation
Authors: Hanqiu Chen, Hang Yang, Stephen BR Fitzmeyer, Cong Hao
Abstract
Implicit Neural Representation (INR) is an innovative approach for representing complex shapes or objects without explicitly defining their geometry or surface structure. Instead, INR represents objects as continuous functions. Previous research has demonstrated the effectiveness of using neural networks as INR for image compression, showcasing comparable performance to traditional methods such as JPEG. However, INR holds potential for various applications beyond image compression. This paper introduces Rapid-INR, a novel approach that utilizes INR for encoding and compressing images, thereby accelerating neural network training in computer vision tasks. Our methodology involves storing the whole dataset directly in INR format on a GPU, mitigating the significant data communication overhead between the CPU and GPU during training. Additionally, the decoding process from INR to RGB format is highly parallelized and executed on-the-fly. To further enhance compression, we propose iterative and dynamic pruning, as well as layer-wise quantization, building upon previous work. We evaluate our framework on the image classification task, utilizing the ResNet-18 backbone network and three commonly used datasets with varying image sizes. Rapid-INR reduces memory consumption to only 5% of the original dataset size and achieves a maximum 6$\times$ speedup over the PyTorch training pipeline, as well as a maximum 1.2x speedup over the DALI training pipeline, with only a marginal decrease in accuracy. Importantly, Rapid-INR can be readily applied to other computer vision tasks and backbone networks with reasonable engineering efforts. Our implementation code is publicly available at https://anonymous.4open.science/r/INR-4BF7.
Keyword: efficient
DNA-TEQ: An Adaptive Exponential Quantization of Tensors for DNN Inference
DenseBAM-GI: Attention Augmented DeneseNet with momentum aided GRU for HMER
Equivariant Neural Networks for Indirect Measurements
Solver algorithm for stabilized space-time formulation of advection-dominated diffusion problem
HNO: Hyena Neural Operator for solving PDEs
CLANet: A Comprehensive Framework for Cross-Batch Cell Line Identification Using Brightfield Images
Envisioning a Next Generation Extended Reality Conferencing System with Efficient Photorealistic Human Rendering
Performance Analysis of Empirical Open-Circuit Voltage Modeling in Lithium Ion Batteries, Part-1: Performance Measures
Performance Analysis of Empirical Open-Circuit Voltage Modeling in Lithium Ion Batteries, Part-2: Data Collection Procedure
An Efficient Sparse Inference Software Accelerator for Transformer-based Language Models on CPUs
GuidedMixup: An Efficient Mixup Strategy Guided by Saliency Maps
Group-based Robustness: A General Framework for Customized Robustness in the Real World
Entanglement Routing over Quantum Networks Using Greenberger-Horne-Zeilinger Measurements
BinaryViT: Pushing Binary Vision Transformers Towards Convolutional Models
Exploring the Representation Power of SPLADE Models
Decomposing spiking neural networks with Graphical Neural Activity Threads
Improving Time and Memory Efficiency of Genetic Algorithms by Storing Populations as Minimum Spanning Trees of Patches
SRL: Scaling Distributed Reinforcement Learning to Over Ten Thousand Cores
Neural Polarizer: A Lightweight and Effective Backdoor Defense via Purifying Poisoned Features
Dynamic-Resolution Model Learning for Object Pile Manipulation
IoT Data Processing for Smart City and Semantic Web Applications
Towards Optimal Randomized Strategies in Adversarial Example Game
Eigensubspace of Temporal-Difference Dynamics and How It Improves Value Approximation in Reinforcement Learning
Unified Language Representation for Question Answering over Text, Tables, and Images
Stop Pre-Training: Adapt Visual-Language Models to Unseen Languages
Algorithms for Computing Maximum Cliques in Hyperbolic Random Graphs
TacMMs: Tactile Mobile Manipulators for Warehouse Automation
Would I have gotten that reward? Long-term credit assignment by counterfactual contribution analysis
Efficient Sobolev approximation of linear parabolic PDEs in high dimensions
Length of Stay prediction for Hospital Management using Domain Adaptation
An Exact Characterisation of Flexibility in Populations of Electric Vehicles
Sampling weights of deep neural networks
Tokenization and the Noiseless Channel
ICDaeLST: Intensity-Controllable Detail Attention-enhanced for Lightweight Fast Style Transfer
Opinion Optimization in Directed Social Networks
SWAT: A System-Wide Approach to Tunable Leakage Mitigation in Encrypted Data Stores
Sustainable Palm Tree Farming: Leveraging IoT and Multi-Modal Data for Early Detection and Mapping of Red Palm Weevil
The ELM Neuron: an Efficient and Expressive Cortical Neuron Model Can Solve Long-Horizon Tasks
MIS-FM: 3D Medical Image Segmentation using Foundation Models Pretrained on a Large-Scale Unannotated Dataset
OSP: Boosting Distributed Model Training with 2-stage Synchronization
The War of the Efficiencies: Understanding the Tension between Carbon and Energy Optimization
Experience Transfer for Robust Direct Data-Driven Control
Diffusion-Jump GNNs: Homophiliation via Learnable Metric Filters
Integrating Large Pre-trained Models into Multimodal Named Entity Recognition with Evidential Fusion
Spatial Reasoning via Deep Vision Models for Robotic Sequential Manipulation
Localized implicit time stepping for the wave equation
On the Predictive Accuracy of Neural Temporal Point Process Models for Continuous-time Event Data
Detect Any Deepfakes: Segment Anything Meets Face Forgery Detection and Localization
RL4CO: an Extensive Reinforcement Learning for Combinatorial Optimization Benchmark
PVP: Personalized Video Prior for Editable Dynamic Portraits using StyleGAN
Ducho: A Unified Framework for the Extraction of Multimodal Features in Recommendation
Generate Anything Anywhere in Any Scene
FogROS2-SGC: A ROS2 Cloud Robotics Platform for Secure Global Connectivity
An Efficient General-Purpose Modular Vision Model via Multi-Task Heterogeneous Training
Keyword: faster
SARC: Soft Actor Retrospective Critic
Momentum Benefits Non-IID Federated Learning Simply and Provably
A systematic study of the foreground-background imbalance problem in deep learning for object detection
Envisioning a Next Generation Extended Reality Conferencing System with Efficient Photorealistic Human Rendering
Learning from Synthetic Human Group Activities
All-intra rate control using low complexity video features for Versatile Video Coding
Sampling weights of deep neural networks
A Formal Perspective on Byte-Pair Encoding
ICDaeLST: Intensity-Controllable Detail Attention-enhanced for Lightweight Fast Style Transfer
SWAT: A System-Wide Approach to Tunable Leakage Mitigation in Encrypted Data Stores
FogROS2-SGC: A ROS2 Cloud Robotics Platform for Secure Global Connectivity
Keyword: mobile
DNA-TEQ: An Adaptive Exponential Quantization of Tensors for DNN Inference
Coordinating Multiple Resources for Optimal Postdisaster Operation of Interdependent Electric Power and Natural Gas Distribution Systems
On the Impact of Interruptions During Multi-Robot Supervision Tasks
Introspective Perception for Mobile Robots
TacMMs: Tactile Mobile Manipulators for Warehouse Automation
A Framework for Securing Health Information Using Blockchain in Cloud Hosted Cyber Physical Systems
FogROS2-SGC: A ROS2 Cloud Robotics Platform for Secure Global Connectivity
Keyword: pruning
An Efficient Sparse Inference Software Accelerator for Transformer-based Language Models on CPUs
NaturalInversion: Data-Free Image Synthesis Improving Real-World Consistency
Rapid-INR: Storage Efficient CPU-free DNN Training Using Implicit Neural Representation
Sparse Model Soups: A Recipe for Improved Pruning via Model Averaging
Can Machines Garden? Systematically Comparing the AlphaGarden vs. Professional Horticulturalists
Keyword: diffusion
Solver algorithm for stabilized space-time formulation of advection-dominated diffusion problem
DiffusionSTR: Diffusion Model for Scene Text Recognition
New twofold saddle-point formulations for Biot poroelasticity with porosity-dependent permeability
Efficient Sobolev approximation of linear parabolic PDEs in high dimensions
SaGess: Sampling Graph Denoising Diffusion Model for Scalable Graph Generation
PFB-Diff: Progressive Feature Blending Diffusion for Text-driven Image Editing
Whole-Body Exploration with a Manipulator Using Heat Equation
One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimization
DreamDiffusion: Generating High-Quality Images from Brain EEG Signals
Diffusion-Jump GNNs: Homophiliation via Learnable Metric Filters
Spiking Denoising Diffusion Probabilistic Models
Learning Structure-Guided Diffusion Model for 2D Human Pose Estimation
Michelangelo: Conditional 3D Shape Generation based on Shape-Image-Text Aligned Latent Representation
ID-Pose: Sparse-view Camera Pose Estimation by Inverting Diffusion Models
Filtered-Guided Diffusion: Fast Filter Guidance for Black-Box Diffusion Models
Generate Anything Anywhere in Any Scene
Keyword: adaptive
DNA-TEQ: An Adaptive Exponential Quantization of Tensors for DNN Inference
Solver algorithm for stabilized space-time formulation of advection-dominated diffusion problem
Deep Equilibrium Multimodal Fusion
NaturalInversion: Data-Free Image Synthesis Improving Real-World Consistency
Dynamic-Resolution Model Learning for Object Pile Manipulation
Metric-aligned Sample Selection and Critical Feature Sampling for Oriented Object Detection
Multi-Scenario Ranking with Adaptive Feature Learning
Influence Maximization based on Simplicial Contagion Model in Hypergraph
RowPress: Amplifying Read Disturbance in Modern DRAM Chips
Re-Rank - Expand - Repeat: Adaptive Query Expansion for Document Retrieval Using Words and Entities
Filtered-Guided Diffusion: Fast Filter Guidance for Black-Box Diffusion Models
Keyword: quantization
DNA-TEQ: An Adaptive Exponential Quantization of Tensors for DNN Inference
Rapid-INR: Storage Efficient CPU-free DNN Training Using Implicit Neural Representation