Abstract
Cloud computing has become the de facto paradigm for delivering software to system users, with organizations and enterprises of all sizes making use of cloud services in some way. On the surface, adopting the cloud appears to be a very efficient approach for offloading concerns such as infrastructure management, logistics, and most importantly for this work, energy consumption and consequent carbon emissions to the cloud service provider. However, this is in many ways not an appropriately accountable solution to managing the contribution of the ICT sector to global emissions. To this effect, in this paper we report on an exploratory case study done in collaboration with a Software as a Service provider operating globally in the telecommunications sector. The study reckons with the service provider using multi-tenant, that is, shared, off-premises data centers for hosting their private cloud infrastructure towards developing a fair model of allocating operational emissions among the service tenants -- customer companies with many distinct users. The developed emissions model has to account for allocating in an appropriate manner the generated emissions between the tenants of the software provider services, and among the different tenants of the same data center. A carbon footprint report generator is developed building on the proposed model which is, in turn, used to present sustainability reports to involved stakeholders for evaluation purposes. Our results show that the model is perceived as transparent, informative, and fair, with requested improvements focusing mainly on the generated reports and the information contained therein.
An Intelligent SDWN Routing Algorithm Based on Network Situational Awareness and Deep Reinforcement Learning
Authors: Jinqiang Li, Miao Ye, Linqiang Huang, Xiaofang Deng, Hongbing Qiu, Yong Wang
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
Due to the highly dynamic changes in wireless network topologies, efficiently obtaining network status information and flexibly forwarding data to improve communication quality of service are important challenges. This article introduces an intelligent routing algorithm (DRL-PPONSA) based on proximal policy optimization deep reinforcement learning with network situational awareness under a software-defined wireless networking architecture. First, a specific data plane is designed for network topology construction and data forwarding. The control plane collects network traffic information, sends flow tables, and uses a GCN-GRU prediction mechanism to perceive future traffic change trends to achieve network situational awareness. Second, a DRL-based data forwarding mechanism is designed in the knowledge plane. The predicted network traffic matrix and topology information matrix are treated as the environment for DRL agents, while next-hop adjacent nodes are treated as executable actions. Accordingly, action selection strategies are designed for different network conditions to achieve more intelligent, flexible, and efficient routing control. The reward function is designed using network link information and various reward and penalty mechanisms. Additionally, importance sampling and gradient clipping techniques are employed during gradient updating to enhance convergence speed and stability. Experimental results show that DRL-PPONSA outperforms traditional routing methods in network throughput, delay, packet loss rate, and wireless node distance. Compared to value-function-based Dueling DQN routing, the convergence speed is significantly improved, and the convergence effect is more stable. Simultaneously, its consumption of hardware storage space is reduced, and efficient routing decisions can be made in real-time using the current network state information.
Optimizing Forest Fire Prevention: Intelligent Scheduling Algorithms for Drone-Based Surveillance System
Abstract
Given the importance of forests and their role in maintaining the ecological balance, which directly affects the planet, the climate, and the life on this planet, this research presents the problem of forest fire monitoring using drones. The forest monitoring process is performed continuously to track any changes in the monitored region within the forest. During fires, drones' capture data is used to increase the follow-up speed and enhance the control process of these fires to prevent their spread. The time factor in such problems determines the success rate of the fire extinguishing process, as appropriate data at the right time may be the decisive factor in controlling fires, preventing their spread, extinguishing them, and limiting their losses. Therefore, this research presented the problem of monitoring task scheduling for drones in the forest monitoring system. This problem is solved by developing several algorithms with the aim of minimizing the total completion time required to carry out all the drones' assigned tasks. System performance is measured by using 990 instances of three different classes. The performed experimental results indicated the effectiveness of the proposed algorithms and their ability to act efficiently to achieve the desired goal. The algorithm $RID$ achieved the best performance with a percentage rate of up to 90.3% with a time of 0.088 seconds.
AnalogNAS: A Neural Network Design Framework for Accurate Inference with Analog In-Memory Computing
Authors: Hadjer Benmeziane, Corey Lammie, Irem Boybat, Malte Rasch, Manuel Le Gallo, Hsinyu Tsai, Ramachandran Muralidhar, Smail Niar, Ouarnoughi Hamza, Vijay Narayanan, Abu Sebastian, Kaoutar El Maghraoui
Abstract
The advancement of Deep Learning (DL) is driven by efficient Deep Neural Network (DNN) design and new hardware accelerators. Current DNN design is primarily tailored for general-purpose use and deployment on commercially viable platforms. Inference at the edge requires low latency, compact and power-efficient models, and must be cost-effective. Digital processors based on typical von Neumann architectures are not conducive to edge AI given the large amounts of required data movement in and out of memory. Conversely, analog/mixed signal in-memory computing hardware accelerators can easily transcend the memory wall of von Neuman architectures when accelerating inference workloads. They offer increased area and power efficiency, which are paramount in edge resource-constrained environments. In this paper, we propose AnalogNAS, a framework for automated DNN design targeting deployment on analog In-Memory Computing (IMC) inference accelerators. We conduct extensive hardware simulations to demonstrate the performance of AnalogNAS on State-Of-The-Art (SOTA) models in terms of accuracy and deployment efficiency on various Tiny Machine Learning (TinyML) tasks. We also present experimental results that show AnalogNAS models achieving higher accuracy than SOTA models when implemented on a 64-core IMC chip based on Phase Change Memory (PCM). The AnalogNAS search code is released: https://github.com/IBM/analog-nas
Statistical Knowledge Assessment for Generative Language Models
Authors: Qingxiu Dong, Jingjing Xu, Lingpeng Kong, Zhifang Sui, Lei Li
Subjects: Computation and Language (cs.CL); Machine Learning (cs.LG)
Abstract
Generative Language Models (GLMs) have demonstrated capabilities to store factual knowledge and answer queries efficiently. Given varying prompts, does a GLM consistently generate factually correct answers? In this paper, we introduce a statistical knowledge assessment framework guided by latent variables and the KaRR metric, which quantifies a model's knowledge by computing its continuous probability across diverse text forms. We conduct a comprehensive comparison of knowledge across 14 GLMs using our framework, including LLaMA, Alpaca, OPT, and others. Our statistical knowledge assessment encompasses 600 relation types and exhibits a strong correlation (0.43 Kendall's $\tau$) with human evaluation. Our findings reveal that the knowledge in GLMs with the same backbone architecture adheres to the scaling law, and that tuning on instruction-following data may compromise the model's ability to generate factually correct text consistently.
Integrated Conflict Management for UAM with Strategic Demand Capacity Balancing and Learning-based Tactical Deconfliction
Authors: Shulu Chen, Antony Evans, Marc Brittain, Peng Wei
Abstract
Urban air mobility (UAM) has the potential to revolutionize our daily transportation, offering rapid and efficient deliveries of passengers and cargo between dedicated locations within and around the urban environment. Before the commercialization and adoption of this emerging transportation mode, however, aviation safety must be guaranteed, i.e., all the aircraft have to be safely separated by strategic and tactical deconfliction. Reinforcement learning has demonstrated effectiveness in the tactical deconfliction of en route commercial air traffic in simulation. However, its performance is found to be dependent on the traffic density. In this project, we propose a novel framework that combines demand capacity balancing (DCB) for strategic conflict management and reinforcement learning for tactical separation. By using DCB to precondition traffic to proper density levels, we show that reinforcement learning can achieve much better performance for tactical safety separation. Our results also indicate that this DCB preconditioning can allow target levels of safety to be met that are otherwise impossible. In addition, combining strategic DCB with reinforcement learning for tactical separation can meet these safety levels while achieving greater operational efficiency than alternative solutions.
The Complexity of Diagonalization
Authors: Nikhil Srivastava
Subjects: Symbolic Computation (cs.SC); Computational Complexity (cs.CC); Data Structures and Algorithms (cs.DS); Numerical Analysis (math.NA)
Abstract
We survey recent progress on efficient algorithms for approximately diagonalizing a square complex matrix in the models of rational (variable precision) and finite (floating point) arithmetic. This question has been studied across several research communities for decades, but many mysteries remain. We present several open problems which we hope will be of broad interest.
MultiPlaneNeRF: Neural Radiance Field with Non-Trainable Representation
Authors: Dominik Zimny, Jacek Tabor, Maciej Zięba, Przemysław Spurek
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
NeRF is a popular model that efficiently represents 3D objects from 2D images. However, vanilla NeRF has a few important limitations. NeRF must be trained on each object separately. The training time is long since we encode the object's shape and color in neural network weights. Moreover, NeRF does not generalize well to unseen data. In this paper, we present MultiPlaneNeRF -- a first model that simultaneously solves all the above problems. Our model works directly on 2D images. We project 3D points on 2D images to produce non-trainable representations. The projection step is not parametrized, and a very shallow decoder can efficiently process the representation. Using existing images as part of NeRF can significantly reduce the number of parameters since we train only a small implicit decoder. Furthermore, we can train MultiPlaneNeRF on a large data set and force our implicit decoder to generalize across many objects. Consequently, we can only replace the 2D images (without additional training) to produce a NeRF representation of the new object. In the experimental section, we demonstrate that MultiPlaneNeRF achieves comparable results to state-of-the-art models for synthesizing new views and has generalization properties.
Unsourced Massive Access-Based Digital Over-the-Air Computation for Efficient Federated Edge Learning
Authors: Li Qiao, Zhen Gao, Zhongxiang Li, Deniz Gündüz
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Abstract
Over-the-air computation (OAC) is a promising technique to achieve fast model aggregation across multiple devices in federated edge learning (FEEL). In addition to the analog schemes, one-bit digital aggregation (OBDA) scheme was proposed to adapt OAC to modern digital wireless systems. However, one-bit quantization in OBDA can result in a serious information loss and slower convergence of FEEL. To overcome this limitation, this paper proposes an unsourced massive access (UMA)-based generalized digital OAC (GD-OAC) scheme. Specifically, at the transmitter, all the devices share the same non-orthogonal UMA codebook for uplink transmission. The local model update of each device is quantized based on the same quantization codebook. Then, each device transmits a sequence selected from the UMA codebook based on the quantized elements of its model update. At the receiver, we propose an approximate message passing-based algorithm for efficient UMA detection and model aggregation. Simulation results show that the proposed GD-OAC scheme significantly accelerates the FEEL convergences compared with the state-of-the-art OBDA scheme while using the same uplink communication resources.
ACRoBat: Optimizing Auto-batching of Dynamic Deep Learning at Compile Time
Authors: Pratik Fegade, Tianqi Chen, Phillip B. Gibbons, Todd C. Mowry
Abstract
Dynamic control flow is an important technique often used to design expressive and efficient deep learning computations for applications such as text parsing, machine translation, exiting early out of deep models and so on. However, the resulting control flow divergence makes batching, an important performance optimization, difficult to perform manually. In this paper, we present ACRoBat, a framework that enables efficient automatic batching for dynamic deep learning computations by performing hybrid static+dynamic compiler optimizations and end-to-end tensor code generation. ACRoBat performs up to 8.5X better than DyNet, a state-of-the-art framework for automatic batching, on an Nvidia GeForce RTX 3070 GPU.
Accelerating MPI Collectives with Process-in-Process-based Multi-object Techniques
Authors: Jiajun Huang, Kaiming Ouyang, Yujia Zhai, Jinyang Liu, Min Si, Ken Raffenetti, Hui Zhou, Atsushi Hori, Zizhong Chen, Yanfei Guo, Rajeev Thakur
Subjects: Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract
In the exascale computing era, optimizing MPI collective performance in high-performance computing (HPC) applications is critical. Current algorithms face performance degradation due to system call overhead, page faults, or data-copy latency, affecting HPC applications' efficiency and scalability. To address these issues, we propose PiP-MColl, a Process-in-Process-based Multi-object Inter-process MPI Collective design that maximizes small message MPI collective performance at scale. PiP-MColl features efficient multiple sender and receiver collective algorithms and leverages Process-in-Process shared memory techniques to eliminate unnecessary system call, page fault overhead, and extra data copy, improving intra- and inter-node message rate and throughput. Our design also boosts performance for larger messages, resulting in comprehensive improvement for various message sizes. Experimental results show that PiP-MColl outperforms popular MPI libraries, including OpenMPI, MVAPICH2, and Intel MPI, by up to 4.6X for MPI collectives like MPI_Scatter and MPI_Allgather.
TSoR: TCP Socket over RDMA Container Network for Cloud Native Computing
Abstract
Cloud-native containerized applications constantly seek high-performance and easy-to-operate container network solutions. RDMA network is a potential enabler with higher throughput and lower latency than the standard TCP/IP network stack. However, several challenges remain in equipping containerized applications with RDMA network: 1) How to deliver transparent improvements without modifying application code; 2) How to integrate RDMA-based network solutions with container orchestration systems; 3) How to efficiently utilize RDMA for container networks. In this paper, we present an RDMA-based container network solution, TCP Socket over RDMA (TSoR), which addresses all the above challenges. To transparently accelerate applications using POSIX socket interfaces without modifications, we integrate TSoR with a container runtime that can intercept system calls for socket interfaces. To be compatible with orchestration systems like Kubernetes, TSoR implements a container network following the Kubernetes network model and satisfies all requirements of the model. To leverage RDMA benefits, TSoR designs a high-performance network stack that efficiently transfers TCP traffic using RDMA network. Thus, TSoR provides a turn-key solution for existing Kubernetes clusters to adopt the high-performance RDMA network with minimal effort. Our evaluation results show that TSoR provides up to 2.3x higher throughput and 64\% lower latency for existing containerized applications, such as Redis key-value store and Node.js web server, with no code changes. TSoR code will be open-sourced.
Smoothing the Landscape Boosts the Signal for SGD: Optimal Sample Complexity for Learning Single Index Models
Authors: Alex Damian, Eshaan Nichani, Rong Ge, Jason D. Lee
Subjects: Machine Learning (cs.LG); Information Theory (cs.IT); Machine Learning (stat.ML)
Abstract
We focus on the task of learning a single index model $\sigma(w^\star \cdot x)$ with respect to the isotropic Gaussian distribution in $d$ dimensions. Prior work has shown that the sample complexity of learning $w^\star$ is governed by the information exponent $k^\star$ of the link function $\sigma$, which is defined as the index of the first nonzero Hermite coefficient of $\sigma$. Ben Arous et al. (2021) showed that $n \gtrsim d^{k^\star-1}$ samples suffice for learning $w^\star$ and that this is tight for online SGD. However, the CSQ lower bound for gradient based methods only shows that $n \gtrsim d^{k^\star/2}$ samples are necessary. In this work, we close the gap between the upper and lower bounds by showing that online SGD on a smoothed loss learns $w^\star$ with $n \gtrsim d^{k^\star/2}$ samples. We also draw connections to statistical analyses of tensor PCA and to the implicit regularization effects of minibatch SGD on empirical losses.
Incremental Causal Graph Learning for Online Unsupervised Root Cause Analysis
Abstract
The task of root cause analysis (RCA) is to identify the root causes of system faults/failures by analyzing system monitoring data. Efficient RCA can greatly accelerate system failure recovery and mitigate system damages or financial losses. However, previous research has mostly focused on developing offline RCA algorithms, which often require manually initiating the RCA process, a significant amount of time and data to train a robust model, and then being retrained from scratch for a new system fault. In this paper, we propose CORAL, a novel online RCA framework that can automatically trigger the RCA process and incrementally update the RCA model. CORAL consists of Trigger Point Detection, Incremental Disentangled Causal Graph Learning, and Network Propagation-based Root Cause Localization. The Trigger Point Detection component aims to detect system state transitions automatically and in near-real-time. To achieve this, we develop an online trigger point detection approach based on multivariate singular spectrum analysis and cumulative sum statistics. To efficiently update the RCA model, we propose an incremental disentangled causal graph learning approach to decouple the state-invariant and state-dependent information. After that, CORAL applies a random walk with restarts to the updated causal graph to accurately identify root causes. The online RCA process terminates when the causal graph and the generated root cause list converge. Extensive experiments on three real-world datasets with case studies demonstrate the effectiveness and superiority of the proposed framework.
PTQD: Accurate Post-Training Quantization for Diffusion Models
Abstract
Diffusion models have recently dominated image synthesis and other related generative tasks. However, the iterative denoising process is expensive in computations at inference time, making diffusion models less practical for low-latency and scalable real-world applications. Post-training quantization of diffusion models can significantly reduce the model size and accelerate the sampling process without requiring any re-training. Nonetheless, applying existing post-training quantization methods directly to low-bit diffusion models can significantly impair the quality of generated samples. Specifically, for each denoising step, quantization noise leads to deviations in the estimated mean and mismatches with the predetermined variance schedule. Moreover, as the sampling process proceeds, the quantization noise may accumulate, resulting in a low signal-to-noise ratio (SNR) in late denoising steps. To address these challenges, we propose a unified formulation for the quantization noise and diffusion perturbed noise in the quantized denoising process. We first disentangle the quantization noise into its correlated and residual uncorrelated parts regarding its full-precision counterpart. The correlated part can be easily corrected by estimating the correlation coefficient. For the uncorrelated part, we calibrate the denoising variance schedule to absorb the excess variance resulting from quantization. Moreover, we propose a mixed-precision scheme to choose the optimal bitwidth for each denoising step, which prefers low bits to accelerate the early denoising steps while high bits maintain the high SNR for the late steps. Extensive experiments demonstrate that our method outperforms previous post-training quantized diffusion models in generating high-quality samples, with only a 0.06 increase in FID score compared to full-precision LDM-4 on ImageNet 256x256, while saving 19.9x bit operations.
Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learning
Authors: Yinglun Xu, Gagandeep Singh
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
Abstract
We propose the first black-box targeted attack against online deep reinforcement learning through reward poisoning during training time. Our attack is applicable to general environments with unknown dynamics learned by unknown algorithms and requires limited attack budgets and computational resources. We leverage a general framework and find conditions to ensure efficient attack under a general assumption of the learning algorithms. We show that our attack is optimal in our framework under the conditions. We experimentally verify that with limited budgets, our attack efficiently leads the learning agent to various target policies under a diverse set of popular DRL environments and state-of-the-art learners.
Zero-Day Backdoor Attack against Text-to-Image Diffusion Models via Personalization
Authors: Yihao Huang, Qing Guo, Felix Juefei-Xu
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Although recent personalization methods have democratized high-resolution image synthesis by enabling swift concept acquisition with minimal examples and lightweight computation, they also present an exploitable avenue for high accessible backdoor attacks. This paper investigates a critical and unexplored aspect of text-to-image (T2I) diffusion models - their potential vulnerability to backdoor attacks via personalization. Our study focuses on a zero-day backdoor vulnerability prevalent in two families of personalization methods, epitomized by Textual Inversion and DreamBooth.Compared to traditional backdoor attacks, our proposed method can facilitate more precise, efficient, and easily accessible attacks with a lower barrier to entry. We provide a comprehensive review of personalization in T2I diffusion models, highlighting the operation and exploitation potential of this backdoor vulnerability. To be specific, by studying the prompt processing of Textual Inversion and DreamBooth, we have devised dedicated backdoor attacks according to the different ways of dealing with unseen tokens and analyzed the influence of triggers and concept images on the attack effect. Our empirical study has shown that the nouveau-token backdoor attack has better attack performance while legacy-token backdoor attack is potentially harder to defend.
Extracting Low-/High- Frequency Knowledge from Graph Neural Networks and Injecting it into MLPs: An Effective GNN-to-MLP Distillation Framework
Authors: Lirong Wu, Haitao Lin, Yufei Huang, Tianyu Fan, Stan Z. Li
Abstract
Recent years have witnessed the great success of Graph Neural Networks (GNNs) in handling graph-related tasks. However, MLPs remain the primary workhorse for practical industrial applications due to their desirable inference efficiency and scalability. To reduce their gaps, one can directly distill knowledge from a well-designed teacher GNN to a student MLP, which is termed as GNN-to-MLP distillation. However, the process of distillation usually entails a loss of information, and ``which knowledge patterns of GNNs are more likely to be left and distilled into MLPs?" becomes an important question. In this paper, we first factorize the knowledge learned by GNNs into low- and high-frequency components in the spectral domain and then derive their correspondence in the spatial domain. Furthermore, we identified a potential information drowning problem for existing GNN-to-MLP distillation, i.e., the high-frequency knowledge of the pre-trained GNNs may be overwhelmed by the low-frequency knowledge during distillation; we have described in detail what it represents, how it arises, what impact it has, and how to deal with it. In this paper, we propose an efficient Full-Frequency GNN-to-MLP (FF-G2M) distillation framework, which extracts both low-frequency and high-frequency knowledge from GNNs and injects it into MLPs. Extensive experiments show that FF-G2M improves over the vanilla MLPs by 12.6% and outperforms its corresponding teacher GNNs by 2.6% averaged over six graph datasets and three common GNN architectures.
OpenShape: Scaling Up 3D Shape Representation Towards Open-World Understanding
Authors: Minghua Liu, Ruoxi Shi, Kaiming Kuang, Yinhao Zhu, Xuanlin Li, Shizhong Han, Hong Cai, Fatih Porikli, Hao Su
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
We introduce OpenShape, a method for learning multi-modal joint representations of text, image, and point clouds. We adopt the commonly used multi-modal contrastive learning framework for representation alignment, but with a specific focus on scaling up 3D representations to enable open-world 3D shape understanding. To achieve this, we scale up training data by ensembling multiple 3D datasets and propose several strategies to automatically filter and enrich noisy text descriptions. We also explore and compare strategies for scaling 3D backbone networks and introduce a novel hard negative mining module for more efficient training. We evaluate OpenShape on zero-shot 3D classification benchmarks and demonstrate its superior capabilities for open-world recognition. Specifically, OpenShape achieves a zero-shot accuracy of 46.8% on the 1,156-category Objaverse-LVIS benchmark, compared to less than 10% for existing methods. OpenShape also achieves an accuracy of 85.3% on ModelNet40, outperforming previous zero-shot baseline methods by 20% and performing on par with some fully-supervised methods. Furthermore, we show that our learned embeddings encode a wide range of visual and semantic concepts (e.g., subcategories, color, shape, style) and facilitate fine-grained text-3D and image-3D interactions. Due to their alignment with CLIP embeddings, our learned shape representations can also be integrated with off-the-shelf CLIP-based models for various applications, such as point cloud captioning and point cloud-conditioned image generation.
Ditto: A Simple and Efficient Approach to Improve Sentence Embeddings
Abstract
Prior studies diagnose the anisotropy problem in sentence representations from pre-trained language models, e.g., BERT, without fine-tuning. Our analysis reveals that the sentence embeddings from BERT suffer from a bias towards uninformative words, limiting the performance in semantic textual similarity (STS) tasks. To address this bias, we propose a simple and efficient unsupervised approach, Diagonal Attention Pooling (Ditto), which weights words with model-based importance estimations and computes the weighted average of word representations from pre-trained models as sentence embeddings. Ditto can be easily applied to any pre-trained language model as a postprocessing operation. Compared to prior sentence embedding approaches, Ditto does not add parameters nor requires any learning. Empirical evaluations demonstrate that our proposed Ditto can alleviate the anisotropy problem and improve various pre-trained models on STS tasks.
Joint BS Mode Selection and Beamforming Design for Cooperative Cell-Free ISAC Networks
Authors: Sifan Liu, Ming Li, Qian Liu
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Abstract
Owing to the promising ability of saving hardware cost and spectrum resources, integrated sensing and communication (ISAC) is regarded as a revolutionary technology for future sixth-generation (6G) networks. The mono-static ISAC systems considered in most of existing works can only obtain limited sensing performance due to the single observation angle and easily blocked transmission links, which motivates researchers to investigate cooperative ISAC networks. In order to further improve the degrees of freedom (DoFs) of cooperative ISAC networks, the transmitter-receiver selection, i.e., BS mode selection problem, is meaningful to be studied. However, to our best knowledge, this crucial problem has not been extensively studied in existing works. In this paper, we consider the joint BS mode selection, transmit beamforming, and receive filter design for cooperative cell-free ISAC networks, where multi-base stations (BSs) cooperatively serve communication users and detect targets. We aim to maximize the sum of sensing signal-to-interference-plus-noise ratio (SINR) under the communication SINR requirements, total power budget, and constraints on the numbers of transmitters and receivers. An efficient joint beamforming design algorithm and three different heuristic BS mode selection methods are proposed to solve this non-convex NP-hard problem. Simulation results demonstrates the advantages of cooperative ISAC networks, the importance of BS mode selection, and the effectiveness of our proposed joint design algorithms.
Two-step Newton's method for deflation-one singular zeros of analytic systems
Abstract
We propose a two-step Newton's method for refining an approximation of a singular zero whose deflation process terminates after one step, also known as a deflation-one singularity. Given an isolated singular zero of a square analytic system, our algorithm exploits an invertible linear operator obtained by combining the Jacobian and a projection of the Hessian in the direction of the kernel of the Jacobian. We prove the quadratic convergence of the two-step Newton method when it is applied to an approximation of a deflation-one singular zero. Also, the algorithm requires a smaller size of matrices than the existing methods, making it more efficient. We demonstrate examples and experiments to show the efficiency of the method.
Adaptive choice of near-optimal expansion points for interpolation-based structure-preserving model reduction
Authors: Quirin Aumann, Steffen W. R. Werner
Subjects: Numerical Analysis (math.NA); Systems and Control (eess.SY); Dynamical Systems (math.DS)
Abstract
Interpolation-based methods are well-established and effective approaches for the efficient generation of accurate reduced-order surrogate models. Common challenges for such methods are the automatic selection of good or even optimal interpolation points and the appropriate size of the reduced-order model. An approach that addresses the first problem for linear, unstructured systems is the Iterative Rational Krylov Algorithm (IRKA), which computes optimal interpolation points through iterative updates by solving linear eigenvalue problems. However, in the case of preserving internal system structures, optimal interpolation points are unknown, and heuristics based on nonlinear eigenvalue problems result in numbers of potential interpolation points that typically exceed the reasonable size of reduced-order systems. In our work, we propose a projection-based iterative interpolation method inspired by IRKA for generally structured systems to adaptively compute near-optimal interpolation points as well as an appropriate size for the reduced-order system. Additionally, the iterative updates of the interpolation points can be chosen such that the reduced-order model provides an accurate approximation in specified frequency ranges of interest. For such applications, our new approach outperforms the established methods in terms of accuracy and computational effort. We show this in numerical examples with different structures.
GraphMoco:a Graph Momentum Contrast Model that Using Multimodel Structure Information for Large-scale Binary Function Representation Learning
Authors: Sun RuiJin, Guo ShiZe, Guo Xi, Pan ZhiSong
Abstract
The ability to compute similarity scores of binary code at the function level is essential for cyber security. A single binary file can contain tens of thousands of functions. A deployable learning framework for cybersecurity applications needs to work not only accurately but also efficiently with large amounts of data. Traditional methods suffer from two drawbacks. First, it is very difficult to annotate different pairs of functions with accurate labels. These supervised learning methods can easily be overtrained with inaccurate labels. The second is that they either use the pre-trained encoder or use the fine-grained graph comparison. However, these methods have shortcomings in terms of time or memory consumption. We focus on large-scale Binary Code Similarity Detection (BCSD) and to mitigate the traditional problems, we propose GraphMoco: a graph momentum contrast model that uses multimodal structure information for large-scale binary function representation learning. We take an unsupervised learning approach and make full use of the structural information in the binary code. It does not require manually labelled similar or dissimilar information. Our models perform efficiently on large amounts of training data. Our experimental results show that our method outperforms the state-of-the-art in terms of accuracy.
Ahead-of-Time P-Tuning
Authors: Daniil Gavrilov, Nikita Balagansky
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Abstract
In this paper, we propose Ahead-of-Time (AoT) P-Tuning, a novel parameter-efficient fine-tuning method for pre-trained Language Models (LMs) that adds input-dependent bias before each Transformer layer. We evaluate AoT P-Tuning on GLUE and SuperGLUE benchmarking datasets using RoBERTa and DeBERTa models, showing that it outperforms BitFit and is comparable or better than other baseline methods for efficient fine-tuning. Additionally, we assess the inference overhead of AoT P-Tuning and demonstrate that it introduces negligible overhead compared to established baseline methods. Our method enables multi-task inference with a single backbone LM, making it a practical solution for real-world applications.
X-IQE: eXplainable Image Quality Evaluation for Text-to-Image Generation with Visual Large Language Models
Authors: Yixiong Chen
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
This paper introduces a novel explainable image quality evaluation approach called X-IQE, which leverages visual large language models (LLMs) to evaluate text-to-image generation methods by generating textual explanations. X-IQE utilizes a hierarchical Chain of Thought (CoT) to enable MiniGPT-4 to produce self-consistent, unbiased texts that are highly correlated with human evaluation. It offers several advantages, including the ability to distinguish between real and generated images, evaluate text-image alignment, and assess image aesthetics without requiring model training or fine-tuning. X-IQE is more cost-effective and efficient compared to human evaluation, while significantly enhancing the transparency and explainability of deep image quality evaluation models. We validate the effectiveness of our method as a benchmark using images generated by prevalent diffusion models. X-IQE demonstrates similar performance to state-of-the-art (SOTA) evaluation methods on COCO Caption, while overcoming the limitations of previous evaluation models on DrawBench, particularly in handling ambiguous generation prompts and text recognition in generated images. Project website: https://github.com/Schuture/Benchmarking-Awesome-Diffusion-Models
Q-SHED: Distributed Optimization at the Edge via Hessian Eigenvectors Quantization
Authors: Nicolò Dal Fabbro, Michele Rossi, Luca Schenato, Subhrakanti Dey
Subjects: Systems and Control (eess.SY); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Optimization and Control (math.OC)
Abstract
Edge networks call for communication efficient (low overhead) and robust distributed optimization (DO) algorithms. These are, in fact, desirable qualities for DO frameworks, such as federated edge learning techniques, in the presence of data and system heterogeneity, and in scenarios where internode communication is the main bottleneck. Although computationally demanding, Newton-type (NT) methods have been recently advocated as enablers of robust convergence rates in challenging DO problems where edge devices have sufficient computational power. Along these lines, in this work we propose Q-SHED, an original NT algorithm for DO featuring a novel bit-allocation scheme based on incremental Hessian eigenvectors quantization. The proposed technique is integrated with the recent SHED algorithm, from which it inherits appealing features like the small number of required Hessian computations, while being bandwidth-versatile at a bit-resolution level. Our empirical evaluation against competing approaches shows that Q-SHED can reduce by up to 60% the number of communication rounds required for convergence.
FLIGHT Mode On: A Feather-Light Network for Low-Light Image Enhancement
Authors: Mustafa Ozcan, Hamza Ergezer, Mustafa Ayazaoglu
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Low-light image enhancement (LLIE) is an ill-posed inverse problem due to the lack of knowledge of the desired image which is obtained under ideal illumination conditions. Low-light conditions give rise to two main issues: a suppressed image histogram and inconsistent relative color distributions with low signal-to-noise ratio. In order to address these problems, we propose a novel approach named FLIGHT-Net using a sequence of neural architecture blocks. The first block regulates illumination conditions through pixel-wise scene dependent illumination adjustment. The output image is produced in the output of the second block, which includes channel attention and denoising sub-blocks. Our highly efficient neural network architecture delivers state-of-the-art performance with only 25K parameters. The method's code, pretrained models and resulting images will be publicly available.
Abstract
This paper introduces EventNet-ITA, a large, multi-domain corpus annotated with event frames for Italian, and presents an efficient approach for multi-label Frame Parsing. The approach is then evaluated on the dataset. Covering a wide range of individual, social and historical phenomena, the main contribution of EventNet-ITA is to provide the research community with a resource for textual event mining and a novel and extensive tool for Frame Parsing in Italian.
Ultra-High Resolution Segmentation with Ultra-Rich Context: A Novel Benchmark
Authors: Deyi Ji, Feng Zhao, Hongtao Lu, Mingyuan Tao, Jieping Ye
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
With the increasing interest and rapid development of methods for Ultra-High Resolution (UHR) segmentation, a large-scale benchmark covering a wide range of scenes with full fine-grained dense annotations is urgently needed to facilitate the field. To this end, the URUR dataset is introduced, in the meaning of Ultra-High Resolution dataset with Ultra-Rich Context. As the name suggests, URUR contains amounts of images with high enough resolution (3,008 images of size 5,120x5,120), a wide range of complex scenes (from 63 cities), rich-enough context (1 million instances with 8 categories) and fine-grained annotations (about 80 billion manually annotated pixels), which is far superior to all the existing UHR datasets including DeepGlobe, Inria Aerial, UDD, etc.. Moreover, we also propose WSDNet, a more efficient and effective framework for UHR segmentation especially with ultra-rich context. Specifically, multi-level Discrete Wavelet Transform (DWT) is naturally integrated to release computation burden while preserve more spatial details, along with a Wavelet Smooth Loss (WSL) to reconstruct original structured context and texture with a smooth constrain. Experiments on several UHR datasets demonstrate its state-of-the-art performance. The dataset is available at https://github.com/jankyee/URUR.
Abstract
Generative modeling has recently undergone remarkable advancements, primarily propelled by the transformative implications of Diffusion Probabilistic Models (DPMs). The impressive capability of these models, however, often entails significant computational overhead during both training and inference. To tackle this challenge, we present Diff-Pruning, an efficient compression method tailored for learning lightweight diffusion models from pre-existing ones, without the need for extensive re-training. The essence of Diff-Pruning is encapsulated in a Taylor expansion over pruned timesteps, a process that disregards non-contributory diffusion steps and ensembles informative gradients to identify important weights. Our empirical assessment, undertaken across four diverse datasets highlights two primary benefits of our proposed method: 1) Efficiency: it enables approximately a 50% reduction in FLOPs at a mere 10% to 20% of the original training expenditure; 2) Consistency: the pruned diffusion models inherently preserve generative behavior congruent with their pre-trained progenitors. Code is available at \url{https://github.com/VainF/Diff-Pruning}.
Unsupervised Pansharpening via Low-rank Diffusion Model
Abstract
Pansharpening is a process of merging a highresolution panchromatic (PAN) image and a low-resolution multispectral (LRMS) image to create a single high-resolution multispectral (HRMS) image. Most of the existing deep learningbased pansharpening methods have poor generalization ability and the traditional model-based pansharpening methods need careful manual exploration for the image structure prior. To alleviate these issues, this paper proposes an unsupervised pansharpening method by combining the diffusion model with the low-rank matrix factorization technique. Specifically, we assume that the HRMS image is decomposed into the product of two low-rank tensors, i.e., the base tensor and the coefficient matrix. The base tensor lies on the image field and has low spectral dimension, we can thus conveniently utilize a pre-trained remote sensing diffusion model to capture its image structures. Additionally, we derive a simple yet quite effective way to preestimate the coefficient matrix from the observed LRMS image, which preserves the spectral information of the HRMS. Extensive experimental results on some benchmark datasets demonstrate that our proposed method performs better than traditional model-based approaches and has better generalization ability than deep learning-based techniques. The code is released in https://github.com/xyrui/PLRDiff.
Lyapunov-Driven Deep Reinforcement Learning for Edge Inference Empowered by Reconfigurable Intelligent Surfaces
Authors: Kyriakos Stylianopoulos, Mattia Merluzzi, Paolo Di Lorenzo, George C. Alexandropoulos
Subjects: Information Theory (cs.IT); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
Abstract
In this paper, we propose a novel algorithm for energy-efficient, low-latency, accurate inference at the wireless edge, in the context of 6G networks endowed with reconfigurable intelligent surfaces (RISs). We consider a scenario where new data are continuously generated/collected by a set of devices and are handled through a dynamic queueing system. Building on the marriage between Lyapunov stochastic optimization and deep reinforcement learning (DRL), we devise a dynamic learning algorithm that jointly optimizes the data compression scheme, the allocation of radio resources (i.e., power, transmission precoding), the computation resources (i.e., CPU cycles), and the RIS reflectivity parameters (i.e., phase shifts), with the aim of performing energy-efficient edge classification with end-to-end (E2E) delay and inference accuracy constraints. The proposed strategy enables dynamic control of the system and of the wireless propagation environment, performing a low-complexity optimization on a per-slot basis while dealing with time-varying radio channels and task arrivals, whose statistics are unknown. Numerical results assess the performance of the proposed RIS-empowered edge inference strategy in terms of trade-off between energy, delay, and accuracy of a classification task.
Abstract
The electronic design industry has undergone a significant transformation, transitioning from traditional hand-drawn designs to modern automated design processes. While Computer-Aided Design (CAD) tools emerged alongside the electronic industry, the current building design process has little to no automation. There is a need for a unified platform to address the complexity of building design and provide a more systematic approach. Platform-based design (PBD), originally developed in the electronic industry, enables efficient design processes by promoting the reuse of hardware and software systems. It also facilitates design space exploration while optimizing performance. This paper proposes a modular approach that divides the building into various disciplines and introduces a design flow using the PBD framework to streamline the design process. We also present a case study that demonstrates the use of the PBD framework in the Heating, Ventilation, and Air Conditioning (HVAC) systems design.
Making More of Little Data: Improving Low-Resource Automatic Speech Recognition Using Data Augmentation
Abstract
The performance of automatic speech recognition (ASR) systems has advanced substantially in recent years, particularly for languages for which a large amount of transcribed speech is available. Unfortunately, for low-resource languages, such as minority languages, regional languages or dialects, ASR performance generally remains much lower. In this study, we investigate whether data augmentation techniques could help improve low-resource ASR performance, focusing on four typologically diverse minority languages or language variants (West Germanic: Gronings, West-Frisian; Malayo-Polynesian: Besemah, Nasal). For all four languages, we examine the use of self-training, where an ASR system trained with the available human-transcribed data is used to generate transcriptions, which are then combined with the original data to train a new ASR system. For Gronings, for which there was a pre-existing text-to-speech (TTS) system available, we also examined the use of TTS to generate ASR training data from text-only sources. We find that using a self-training approach consistently yields improved performance (a relative WER reduction up to 20.5% compared to using an ASR system trained on 24 minutes of manually transcribed speech). The performance gain from TTS augmentation for Gronings was even stronger (up to 25.5% relative reduction in WER compared to a system based on 24 minutes of manually transcribed speech). In sum, our results show the benefit of using self-training or (if possible) TTS-generated data as an efficient solution to overcome the limitations of data availability for resource-scarce languages in order to improve ASR performance.
Learning Activation Functions for Sparse Neural Networks
Authors: Mohammad Loni, Aditya Mohan, Mehdi Asadi, Marius Lindauer
Subjects: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
Abstract
Sparse Neural Networks (SNNs) can potentially demonstrate similar performance to their dense counterparts while saving significant energy and memory at inference. However, the accuracy drop incurred by SNNs, especially at high pruning ratios, can be an issue in critical deployment conditions. While recent works mitigate this issue through sophisticated pruning techniques, we shift our focus to an overlooked factor: hyperparameters and activation functions. Our analyses have shown that the accuracy drop can additionally be attributed to (i) Using ReLU as the default choice for activation functions unanimously, and (ii) Fine-tuning SNNs with the same hyperparameters as dense counterparts. Thus, we focus on learning a novel way to tune activation functions for sparse networks and combining these with a separate hyperparameter optimization (HPO) regime for sparse networks. By conducting experiments on popular DNN models (LeNet-5, VGG-16, ResNet-18, and EfficientNet-B0) trained on MNIST, CIFAR-10, and ImageNet-16 datasets, we show that the novel combination of these two approaches, dubbed Sparse Activation Function Search, short: SAFS, results in up to 15.53%, 8.88%, and 6.33% absolute improvement in the accuracy for LeNet-5, VGG-16, and ResNet-18 over the default training protocols, especially at high pruning ratios. Our code can be found at https://github.com/automl/SAFS
Stopping Criteria for the Conjugate Gradient Algorithm in High-Order Finite Element Methods
Authors: Yichen Guo, Eric de Sturler, Tim Warburton
Abstract
We introduce three new stopping criteria that balance algebraic and discretization errors for the conjugate gradient algorithm applied to high-order finite element discretizations of Poisson problems. The current state of the art stopping criteria compare a posteriori estimates of discretization error against estimates of the algebraic error. Firstly, we propose a new error indicator derived from a recovery-based error estimator that is less computationally expensive and more reliable. Secondly, we introduce a new stopping criterion that suggests stopping when the norm of the linear residual is less than a small fraction of an error indicator derived directly from the residual. This indicator shares the same mesh size and polynomial degree scaling as the norm of the residual, resulting in a robust criterion regardless of the mesh size, the polynomial degree, and the shape regularity of the mesh. Thirdly, in solving Poisson problems with highly variable piecewise constant coefficients, we introduce a subdomain-based criterion that recommends stopping when the norm of the linear residual restricted to each subdomain is smaller than the corresponding indicator also restricted to that subdomain. Numerical experiments, including tests with anisotropic meshes and highly variable piecewise constant coefficients, demonstrate that the proposed criteria efficiently avoid both premature termination and over-solving.
Near-Field 3D Localization via MIMO Radar: Cramér-Rao Bound and Estimator Design
Authors: Haocheng Hua, Jie Xu
Subjects: Information Theory (cs.IT); Signal Processing (eess.SP)
Abstract
Future sixth-generation (6G) networks are envisioned to provide both sensing and communications functionalities by using densely deployed base stations (BSs) with massive antennas operating in millimeter wave (mmWave) and terahertz (THz). Due to the large number of antennas and the high frequency band, the sensing and communications will operate within the near-field region, thus making the conventional designs based on the far-field channel models inapplicable. This paper studies a near-field multiple-input-multiple-output (MIMO) radar sensing system, in which the transceivers with massive antennas aim to localize multiple near-field targets in the three-dimensional (3D) space. In particular, we adopt a general wavefront propagation model by considering the exact spherical wavefront with both channel phase and amplitude variations over different antennas. Besides, we consider the general transmit signal waveforms and also consider the unknown cluttered environments. Under this setup, the unknown parameters to estimate include the 3D coordinates and the complex reflection coefficients of the multiple targets, as well as the noise and interference covariance matrix. Accordingly, we derive the Cram\'er-Rao bound (CRB) for estimating the target coordinates and reflection coefficients. Next, to facilitate practical localization, we propose an efficient estimator based on the 3D approximate cyclic optimization (3D-ACO), which is obtained following the maximum likelihood (ML) criterion. Finally, numerical results show that considering the exact antenna-varying channel amplitudes achieves more accurate CRB as compared to prior works based on constant channel amplitudes across antennas, especially when the targets are close to the transceivers. It is also shown that the proposed estimator achieves localization performance close to the derived CRB, thus validating its superior performance.
How does the task complexity of masked pretraining objectives affect downstream performance?
Abstract
Masked language modeling (MLM) is a widely used self-supervised pretraining objective, where a model needs to predict an original token that is replaced with a mask given contexts. Although simpler and computationally efficient pretraining objectives, e.g., predicting the first character of a masked token, have recently shown comparable results to MLM, no objectives with a masking scheme actually outperform it in downstream tasks. Motivated by the assumption that their lack of complexity plays a vital role in the degradation, we validate whether more complex masked objectives can achieve better results and investigate how much complexity they should have to perform comparably to MLM. Our results using GLUE, SQuAD, and Universal Dependencies benchmarks demonstrate that more complicated objectives tend to show better downstream results with at least half of the MLM complexity needed to perform comparably to MLM. Finally, we discuss how we should pretrain a model using a masked objective from the task complexity perspective.
Mode Connectivity in Auction Design
Authors: Christoph Hertrich, Yixin Tao, László A. Végh
Subjects: Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE); Machine Learning (stat.ML)
Abstract
Optimal auction design is a fundamental problem in algorithmic game theory. This problem is notoriously difficult already in very simple settings. Recent work in differentiable economics showed that neural networks can efficiently learn known optimal auction mechanisms and discover interesting new ones. In an attempt to theoretically justify their empirical success, we focus on one of the first such networks, RochetNet, and a generalized version for affine maximizer auctions. We prove that they satisfy mode connectivity, i.e., locally optimal solutions are connected by a simple, piecewise linear path such that every solution on the path is almost as good as one of the two local optima. Mode connectivity has been recently investigated as an intriguing empirical and theoretically justifiable property of neural networks used for prediction problems. Our results give the first such analysis in the context of differentiable economics, where neural networks are used directly for solving non-convex optimization problems.
The Dilemma of Choice: Addressing Constraint Selection for Autonomous Robotic Agents
Abstract
The tasks that an autonomous agent is expected to perform are often optional or are incompatible with each other owing to the agent's limited actuation capabilities, specifically the dynamics and control input bounds. We encode tasks as time-dependent state constraints and leverage the advances in multi-objective optimization to formulate the problem of choosing tasks as selection of a feasible subset of constraints that can be satisfied for all time and maximizes a performance metric. We show that this problem, although amenable to reachability or mixed integer model predictive control-based analysis in the offline phase, is NP-Hard in general and therefore requires heuristics to be solved efficiently. When incompatibility in constraints is observed under a given policy that imposes task constraints at each time step in an optimization problem, we assign a Lagrange score to each of these constraints based on the variation in the corresponding Lagrange multipliers over the compatible time horizon. These scores are then used to decide the order in which constraints are dropped in a greedy strategy. We further employ a genetic algorithm to improve upon the greedy strategy. We evaluate our method on a robot waypoint following task when the low-level controllers that impose state constraints are described by Control Barrier Function-based Quadratic Programs and provide a comparison with waypoint selection based on knowledge of backward reachable sets.
Generalized Planning in PDDL Domains with Pretrained Large Language Models
Authors: Tom Silver, Soham Dan, Kavitha Srinivas, Joshua B. Tenenbaum, Leslie Pack Kaelbling, Michael Katz
Abstract
Recent work has considered whether large language models (LLMs) can function as planners: given a task, generate a plan. We investigate whether LLMs can serve as generalized planners: given a domain and training tasks, generate a program that efficiently produces plans for other tasks in the domain. In particular, we consider PDDL domains and use GPT-4 to synthesize Python programs. We also consider (1) Chain-of-Thought (CoT) summarization, where the LLM is prompted to summarize the domain and propose a strategy in words before synthesizing the program; and (2) automated debugging, where the program is validated with respect to the training tasks, and in case of errors, the LLM is re-prompted with four types of feedback. We evaluate this approach in seven PDDL domains and compare it to four ablations and four baselines. Overall, we find that GPT-4 is a surprisingly powerful generalized planner. We also conclude that automated debugging is very important, that CoT summarization has non-uniform impact, that GPT-4 is far superior to GPT-3.5, and that just two training tasks are often sufficient for strong generalization.
Optimistic Natural Policy Gradient: a Simple Efficient Policy Optimization Framework for Online RL
Authors: Qinghua Liu, Gellért Weisz, András György, Chi Jin, Csaba Szepesvári
Abstract
While policy optimization algorithms have played an important role in recent empirical success of Reinforcement Learning (RL), the existing theoretical understanding of policy optimization remains rather limited -- they are either restricted to tabular MDPs or suffer from highly suboptimal sample complexity, especial in online RL where exploration is necessary. This paper proposes a simple efficient policy optimization framework -- Optimistic NPG for online RL. Optimistic NPG can be viewed as simply combining of the classic natural policy gradient (NPG) algorithm [Kakade, 2001] with optimistic policy evaluation subroutines to encourage exploration. For $d$-dimensional linear MDPs, Optimistic NPG is computationally efficient, and learns an $\varepsilon$-optimal policy within $\tilde{O}(d^2/\varepsilon^3)$ samples, which is the first computationally efficient algorithm whose sample complexity has the optimal dimension dependence $\tilde{\Theta}(d^2)$. It also improves over state-of-the-art results of policy optimization algorithms [Zanette et al., 2021] by a factor of $d$. For general function approximation that subsumes linear MDPs, Optimistic NPG, to our best knowledge, is also the first policy optimization algorithm that achieves the polynomial sample complexity for learning near-optimal policies.
SPSQL: Step-by-step Parsing Based Framework for Text-to-SQL Generation
Authors: Ran Shen, Gang Sun, Hao Shen, Yiling Li, Liangfeng Jin, Han Jiang
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Databases (cs.DB)
Abstract
Converting text into the structured query language (Text2SQL) is a research hotspot in the field of natural language processing (NLP), which has broad application prospects. In the era of big data, the use of databases has penetrated all walks of life, in which the collected data is large in scale, diverse in variety, and wide in scope, making the data query cumbersome and inefficient, and putting forward higher requirements for the Text2SQL model. In practical applications, the current mainstream end-to-end Text2SQL model is not only difficult to build due to its complex structure and high requirements for training data, but also difficult to adjust due to massive parameters. In addition, the accuracy of the model is hard to achieve the desired result. Based on this, this paper proposes a pipelined Text2SQL method: SPSQL. This method disassembles the Text2SQL task into four subtasks--table selection, column selection, SQL generation, and value filling, which can be converted into a text classification problem, a sequence labeling problem, and two text generation problems, respectively. Then, we construct data formats of different subtasks based on existing data and improve the accuracy of the overall model by improving the accuracy of each submodel. We also use the named entity recognition module and data augmentation to optimize the overall model. We construct the dataset based on the marketing business data of the State Grid Corporation of China. Experiments demonstrate our proposed method achieves the best performance compared with the end-to-end method and other pipeline methods.
Blendstrings: an environment for computing with smooth functions
Abstract
A "blendstring" is a piecewise polynomial interpolant with high-degree two-point Hermite interpolational polynomials on each piece, analogous to a cubic spline. Blendstrings are smoother and can be more accurate than cubic splines, and can be used to represent smooth functions on a line segment or polygonal path in the complex plane. I sketch some properties of blendstrings, including efficient methods for evaluation, differentiation, and integration, as well as a prototype Maple implementation. Blendstrings can be differentiated and integrated exactly and can be combined algebraically. I also show applications of blendstrings to solving differential equations and computing Mathieu functions and generalized Mathieu eigenfunctions.
Blockwise inversion and algorithms for inverting large partitioned matrices
Authors: R. Thiru Senthil
Subjects: Numerical Analysis (math.NA); High Energy Physics - Experiment (hep-ex); High Energy Physics - Phenomenology (hep-ph); Mathematical Physics (math-ph)
Abstract
Using the blockwise matrix inversion, inversions of large matrices with different ways of memory handling are presented in this article. Algorithm for performing inversion of matrix which is partitioned into large number of blocks is presented in which inversions and multiplications involving the blocks can be carried out with parallel processing. Optimized memory handling and efficient methods for intermediate multiplications among the partitioned blocks are implemented in this algorithm.
Using Symbolic Computation to Analyze Zero-Hopf Bifurcations of Polynomial Differential Systems
Authors: Bo Huang
Subjects: Symbolic Computation (cs.SC); Dynamical Systems (math.DS)
Abstract
This paper is devoted to the study of infinitesimal limit cycles that can bifurcate from zero-Hopf equilibria of differential systems based on the averaging method. We develop an efficient symbolic program using Maple for computing the averaged functions of any order for continuous differential systems in arbitrary dimension. The program allows us to systematically analyze zero-Hopf bifurcations of polynomial differential systems using symbolic computation methods. We show that for the first-order averaging, $\ell\in{0,1,\ldots,2^{n-3}}$ limit cycles can bifurcate from the zero-Hopf equilibrium for the general class of perturbed differential systems and up to the second-order averaging, the maximum number of limit cycles can be determined by computing the mixed volume of a polynomial system obtained from the averaged functions. A number of examples are presented to demonstrate the effectiveness of the proposed algorithmic approach.
mLongT5: A Multilingual and Efficient Text-To-Text Transformer for Longer Sequences
Authors: David Uthus, Santiago Ontañón, Joshua Ainslie, Mandy Guo
Abstract
We present our work on developing a multilingual, efficient text-to-text transformer that is suitable for handling long inputs. This model, called mLongT5, builds upon the architecture of LongT5, while leveraging the multilingual datasets used for pretraining mT5 and the pretraining tasks of UL2. We evaluate this model on a variety of multilingual summarization and question-answering tasks, and the results show stronger performance for mLongT5 when compared to existing multilingual models such as mBART or M-BERT.
SimOAP: Improve Coherence and Consistency in Persona-based Dialogue Generation via Over-sampling and Post-evaluation
Abstract
Language models trained on large-scale corpora can generate remarkably fluent results in open-domain dialogue. However, for the persona-based dialogue generation task, consistency and coherence are also key factors, which are great challenges for language models. Existing works mainly focus on valuable data filtering, model structure modifying, or objective function designing, while their improvements are limited and hard to generalize to all types of pre-trained language models. However, we find that language models can produce consistent and coherent responses if we consider enough generations. Thus, the problems lay in large-scale response generation and target response selection. In this work, a simple but effective two-stage SimOAP strategy is proposed, i.e., over-sampling and post-evaluation. The over-sampling stage takes large-scale responses from existing trained models efficiently via off-the-shelf distilling and compressing methods, and the post-evaluation stage selects a good response based on multiple well-designed evaluation metrics from large-scale candidates. Experimental results show that the proposed plug-in SimOAP strategy improves the backbone models and outperforms the baseline strategies in both automatic and human evaluations.
Convergence Analysis of Over-the-Air FL with Compression and Power Control via Clipping
Authors: Haifeng Wen, Hong Xing, Osvaldo Simeone
Subjects: Information Theory (cs.IT); Machine Learning (cs.LG); Signal Processing (eess.SP)
Abstract
One of the key challenges towards the deployment of over-the-air federated learning (AirFL) is the design of mechanisms that can comply with the power and bandwidth constraints of the shared channel, while causing minimum deterioration to the learning performance as compared to baseline noiseless implementations. For additive white Gaussian noise (AWGN) channels with instantaneous per-device power constraints, prior work has demonstrated the optimality of a power control mechanism based on norm clipping. This was done through the minimization of an upper bound on the optimality gap for smooth learning objectives satisfying the Polyak-{\L}ojasiewicz (PL) condition. In this paper, we make two contributions to the development of AirFL based on norm clipping, which we refer to as AirFL-Clip. First, we provide a convergence bound for AirFLClip that applies to general smooth and non-convex learning objectives. Unlike existing results, the derived bound is free from run-specific parameters, thus supporting an offline evaluation. Second, we extend AirFL-Clip to include Top-k sparsification and linear compression. For this generalized protocol, referred to as AirFL-Clip-Comp, we derive a convergence bound for general smooth and non-convex learning objectives. We argue, and demonstrate via experiments, that the only time-varying quantities present in the bound can be efficiently estimated offline by leveraging the well-studied properties of sparse recovery algorithms.
Exploring the Carbon Footprint of Hugging Face's ML Models: A Repository Mining Study
Authors: Joel Castaño, Silverio Martínez-Fernández, Xavier Franch, Justus Bogner
Subjects: Machine Learning (cs.LG); Computers and Society (cs.CY); Information Retrieval (cs.IR); Machine Learning (stat.ML)
Abstract
The rise of machine learning (ML) systems has exacerbated their carbon footprint due to increased capabilities and model sizes. However, there is scarce knowledge on how the carbon footprint of ML models is actually measured, reported, and evaluated. In light of this, the paper aims to analyze the measurement of the carbon footprint of 1,417 ML models and associated datasets on Hugging Face, which is the most popular repository for pretrained ML models. The goal is to provide insights and recommendations on how to report and optimize the carbon efficiency of ML models. The study includes the first repository mining study on the Hugging Face Hub API on carbon emissions. This study seeks to answer two research questions: (1) how do ML model creators measure and report carbon emissions on Hugging Face Hub?, and (2) what aspects impact the carbon emissions of training ML models? The study yielded several key findings. These include a decreasing proportion of carbon emissions-reporting models, a slight decrease in reported carbon footprint on Hugging Face over the past 2 years, and a continued dominance of NLP as the main application domain. Furthermore, the study uncovers correlations between carbon emissions and various attributes such as model size, dataset size, and ML application domains. These results highlight the need for software measurements to improve energy reporting practices and promote carbon-efficient model development within the Hugging Face community. In response to this issue, two classifications are proposed: one for categorizing models based on their carbon emission reporting practices and another for their carbon efficiency. The aim of these classification proposals is to foster transparency and sustainable model development within the ML community.
Efficient Prompting via Dynamic In-Context Learning
Authors: Wangchunshu Zhou, Yuchen Eleanor Jiang, Ryan Cotterell, Mrinmaya Sachan
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
The primary way of building AI applications is shifting from training specialist models to prompting generalist models. A common practice for prompting generalist models, often referred to as in-context learning, is to append a few examples (demonstrations) to the prompt to help the model better understand the task. While effective, in-context learning can be inefficient because it makes the input prompt much longer, consuming valuable space in the context window and leading to larger computational costs. In this paper, we propose DynaICL, a recipe for efficient prompting with black-box generalist models that dynamically allocate in-context examples according to the input complexity and the computational budget. To achieve this, we train a meta controller that predicts the number of in-context examples suitable for the generalist model to make a good prediction based on the performance-efficiency trade-off for a specific input. We then dynamically allocate the number of demonstrations for an input according to predictions from the meta controller and the given computation budget. Experimental results show that dynamic example allocation helps achieve a better performance-efficiency trade-off in two practical settings where computational resources or the required performance is constrained. Specifically, DynaICL saves up to 46% token budget compared to the common practice that allocates the same number of in-context examples to each input. We also find that a meta controller trained on a certain backbone model and tasks can successfully generalize to unseen models and tasks.
Keyword: faster
RAMP: Hierarchical Reactive Motion Planning for Manipulation Tasks Using Implicit Signed Distance Functions
Abstract
We introduce Reactive Action and Motion Planner (RAMP), which combines the strengths of search-based and reactive approaches for motion planning. In essence, RAMP is a hierarchical approach where a novel variant of a Model Predictive Path Integral (MPPI) controller is used to generate trajectories which are then followed asynchronously by a local vector field controller. We demonstrate, in the context of a table clearing application, that RAMP can rapidly find paths in the robot's configuration space, satisfy task and robot-specific constraints, and provide safety by reacting to static or dynamically moving obstacles. RAMP achieves superior performance through a number of key innovations: we use Signed Distance Function (SDF) representations directly from the robot configuration space, both for collision checking and reactive control. The use of SDFs allows for a smoother definition of collision cost when planning for a trajectory, and is critical in ensuring safety while following trajectories. In addition, we introduce a novel variant of MPPI which, combined with the safety guarantees of the vector field trajectory follower, performs incremental real-time global trajectory planning. Simulation results establish that our method can generate paths that are comparable to traditional and state-of-the-art approaches in terms of total trajectory length while being up to 30 times faster. Real-world experiments demonstrate the safety and effectiveness of our approach in challenging table clearing scenarios.
Democratized Diffusion Language Model
Authors: Nikita Balagansky, Daniil Gavrilov
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Abstract
Despite the potential benefits of Diffusion Models for NLP applications, publicly available implementations, trained models, or reproducible training procedures currently need to be publicly available. We present the Democratized Diffusion Language Model (DDLM), based on the Continuous Diffusion for Categorical Data (CDCD) framework, to address these challenges. We propose a simplified training procedure for DDLM using the C4 dataset and perform an in-depth analysis of the trained model's behavior. Furthermore, we introduce a novel early-exiting strategy for faster sampling with models trained with score interpolation. Since no previous works aimed at solving downstream tasks with pre-trained Diffusion LM (e.g., classification tasks), we experimented with GLUE Benchmark to study the ability of DDLM to transfer knowledge. With this paper, we propose available training and evaluation pipelines to other researchers and pre-trained DDLM models, which could be used in future research with Diffusion LMs.
A Lexical-aware Non-autoregressive Transformer-based ASR Model
Authors: Chong-En Lin, Kuan-Yu Chen
Subjects: Computation and Language (cs.CL); Sound (cs.SD); Audio and Speech Processing (eess.AS)
Abstract
Non-autoregressive automatic speech recognition (ASR) has become a mainstream of ASR modeling because of its fast decoding speed and satisfactory result. To further boost the performance, relaxing the conditional independence assumption and cascading large-scaled pre-trained models are two active research directions. In addition to these strategies, we propose a lexical-aware non-autoregressive Transformer-based (LA-NAT) ASR framework, which consists of an acoustic encoder, a speech-text shared encoder, and a speech-text shared decoder. The acoustic encoder is used to process the input speech features as usual, and the speech-text shared encoder and decoder are designed to train speech and text data simultaneously. By doing so, LA-NAT aims to make the ASR model aware of lexical information, so the resulting model is expected to achieve better results by leveraging the learned linguistic knowledge. A series of experiments are conducted on the AISHELL-1, CSJ, and TEDLIUM 2 datasets. According to the experiments, the proposed LA-NAT can provide superior results than other recently proposed non-autoregressive ASR models. In addition, LA-NAT is a relatively compact model than most non-autoregressive ASR models, and it is about 58 times faster than the classic autoregressive model.
TAPIR: Learning Adaptive Revision for Incremental Natural Language Understanding with a Two-Pass Model
Authors: Patrick Kahardipraja, Brielen Madureira, David Schlangen
Abstract
Language is by its very nature incremental in how it is produced and processed. This property can be exploited by NLP systems to produce fast responses, which has been shown to be beneficial for real-time interactive applications. Recent neural network-based approaches for incremental processing mainly use RNNs or Transformers. RNNs are fast but monotonic (cannot correct earlier output, which can be necessary in incremental processing). Transformers, on the other hand, consume whole sequences, and hence are by nature non-incremental. A restart-incremental interface that repeatedly passes longer input prefixes can be used to obtain partial outputs, while providing the ability to revise. However, this method becomes costly as the sentence grows longer. In this work, we propose the Two-pass model for AdaPtIve Revision (TAPIR) and introduce a method to obtain an incremental supervision signal for learning an adaptive revision policy. Experimental results on sequence labelling show that our model has better incremental performance and faster inference speed compared to restart-incremental Transformers, while showing little degradation on full sequences.
Abstract
In this work, we present a new benchmarking suite with new real-life inspired skewed workloads to test the performance of concurrent index data structures. We started this project to prepare workloads specifically for self-adjusting data structures, i.e., they handle more frequent requests faster, and, thus, should perform better than their standard counterparts. We looked over the commonly used suites to test performance of concurrent indices trying to find an inspiration: Synchrobench, Setbench, YCSB, and TPC - and we found several issues with them. The major problem is that they are not flexible: it is difficult to introduce new workloads, it is difficult to set the duration of the experiments, and it is difficult to change the parameters. We decided to solve this issue by presenting a new suite based on Synchrobench. Finally, we highlight the problem of measuring performance of data structures. We show that the relative performance of data structures highly depends on the workload: it is not clear which data structure is best. For that, we take three state-of-the-art concurrent binary search trees and run them on the workloads from our benchmarking suite. As a result, we get six experiments with all possible relative performance of the chosen data structures.
Hibernate Container: A Deflated Container Mode for Fast Startup and High-density Deployment in Serverless Computing
Abstract
Serverless computing is a popular cloud computing paradigm, which requires low response latency to handle on-demand user requests. There are two prominent techniques employed for reducing the response latency: keep fully initialized containers alive (Warm Container) or reduce the new container startup (cold start) latency. This paper presents the 3rd container startup mode: Hibernate Container, which starts faster than the cold start container mode and consumes less memory than the Warm Container mode. Hibernate Container is essentially a "deflated" Warm Container. Its application memory is swapped out to disk, the freed memory is reclaimed and file based mmap memory is cleaned-up. The Hibernate Container's deflated memory is inflated in response to user requests. As Hibernate Container's application is fully initialized, its response latency is less than the cold start mode; and as the application memory is deflated, its memory consumption is less than the Warm Container mode. Additionally, when a Hibernate Container is "woken up" to process a request, the Woken-up Container has similar response latency to Warm Container but less memory consumption because not all the deflated memory needs to be inflated. We implemented the Hibernate technique as part of the open source Quark secure container runtime project and our test demonstrated that Hibernate Container consumes about 7\% to 25\% of the Warm Container memory. All of this results in a higher deployment density, lower latency and appreciable improvements in the overall system performance.
Deep PackGen: A Deep Reinforcement Learning Framework for Adversarial Network Packet Generation
Authors: Soumyadeep Hore, Jalal Ghadermazi, Diwas Paudel, Ankit Shah, Tapas K. Das, Nathaniel D. Bastian
Subjects: Cryptography and Security (cs.CR); Machine Learning (cs.LG)
Abstract
Recent advancements in artificial intelligence (AI) and machine learning (ML) algorithms, coupled with the availability of faster computing infrastructure, have enhanced the security posture of cybersecurity operations centers (defenders) through the development of ML-aided network intrusion detection systems (NIDS). Concurrently, the abilities of adversaries to evade security have also increased with the support of AI/ML models. Therefore, defenders need to proactively prepare for evasion attacks that exploit the detection mechanisms of NIDS. Recent studies have found that the perturbation of flow-based and packet-based features can deceive ML models, but these approaches have limitations. Perturbations made to the flow-based features are difficult to reverse-engineer, while samples generated with perturbations to the packet-based features are not playable. Our methodological framework, Deep PackGen, employs deep reinforcement learning to generate adversarial packets and aims to overcome the limitations of approaches in the literature. By taking raw malicious network packets as inputs and systematically making perturbations on them, Deep PackGen camouflages them as benign packets while still maintaining their functionality. In our experiments, using publicly available data, Deep PackGen achieved an average adversarial success rate of 66.4\% against various ML models and across different attack types. Our investigation also revealed that more than 45\% of the successful adversarial samples were out-of-distribution packets that evaded the decision boundaries of the classifiers. The knowledge gained from our study on the adversary's ability to make specific evasive perturbations to different types of malicious packets can help defenders enhance the robustness of their NIDS against evolving adversarial attacks.
MVPSNet: Fast Generalizable Multi-view Photometric Stereo
Abstract
We propose a fast and generalizable solution to Multi-view Photometric Stereo (MVPS), called MVPSNet. The key to our approach is a feature extraction network that effectively combines images from the same view captured under multiple lighting conditions to extract geometric features from shading cues for stereo matching. We demonstrate these features, termed `Light Aggregated Feature Maps' (LAFM), are effective for feature matching even in textureless regions, where traditional multi-view stereo methods fail. Our method produces similar reconstruction results to PS-NeRF, a state-of-the-art MVPS method that optimizes a neural network per-scene, while being 411$\times$ faster (105 seconds vs. 12 hours) in inference. Additionally, we introduce a new synthetic dataset for MVPS, sMVPS, which is shown to be effective to train a generalizable MVPS method.
Keyword: mobile
Sim-MEES: Modular End-Effector System Grasping Dataset for Mobile Manipulators in Cluttered Environments
Abstract
In this paper, we present Sim-MEES: a large-scale synthetic dataset that contains 1,550 objects with varying difficulty levels and physics properties, as well as 11 million grasp labels for mobile manipulators to plan grasps using different gripper modalities in cluttered environments. Our dataset generation process combines analytic models and dynamic simulations of the entire cluttered environment to provide accurate grasp labels. We provide a detailed study of our proposed labeling process for both parallel jaw grippers and suction cup grippers, comparing them with state-of-the-art methods to demonstrate how Sim-MEES can provide precise grasp labels in cluttered environments.
Multi-microservice migration modelling, comparison, and potential in 5G/6G mobile edge computing: A non-average parameter values approach
Authors: Arshin Rezazadeh, Hanan Lutfiyya
Subjects: Networking and Internet Architecture (cs.NI); Distributed, Parallel, and Cluster Computing (cs.DC)
Abstract
Cloud, fog, and edge computing integration with future mobile Internet-of-Things (IoT) devices and related applications in 5G/6G networks will become more practical in the coming years. Containers became the de facto virtualization technique that replaced Virtual Memory (VM). Mobile IoT applications, e.g., intelligent transportation and augmented reality, incorporating fog-edge, have increased the demand for a millisecond-scale response and processing time. Edge Computing reduces remote network traffic and latency. These services must run on edge nodes that are physically close to devices. However, classical migration techniques may not meet the requirements of future mission-critical IoT applications. IoT mobile devices have limited resources for running multiple services, and client-server latency worsens when fog-edge services must migrate to maintain proximity in light of device mobility. This study analyzes the performance of the MiGrror migration method and the pre-copy live migration method when the migration of multiple VMs/containers is considered. This paper presents mathematical models for the stated methods and provides migration guidelines and comparisons for services to be implemented as multiple containers, as in microservice-based environments. Experiments demonstrate that MiGrror outperforms the pre-copy technique and, unlike conventional live migrations, can maintain less than 10 milliseconds of downtime and reduce migration time with a minimal bandwidth overhead. The results show that MiGrror can improve service continuity and availability for users. Most significant is that the model can use average and non-average values for different parameters during migration to achieve improved and more accurate results, while other research typically only uses average values. This paper shows that using only average parameter values in migration can lead to inaccurate results.
Abstract
We present the first controller for quasistatic robotic planar pushing with single-point contact using only force feedback. We consider a mobile robot equipped with a force-torque sensor to measure the force at the contact point with the pushed object (the "slider"). The parameters of the slider are not known to the controller, nor is feedback on the slider's pose. We assume that the global position of the contact point is always known and that the approximate initial position of the slider is provided. We focus specifically on the case when it is desired to push the slider along a straight line. Simulations and real-world experiments show that our controller yields stable pushes that are robust to a wide range of slider parameters and state perturbations.
XFormer: Fast and Accurate Monocular 3D Body Capture
Abstract
We present XFormer, a novel human mesh and motion capture method that achieves real-time performance on consumer CPUs given only monocular images as input. The proposed network architecture contains two branches: a keypoint branch that estimates 3D human mesh vertices given 2D keypoints, and an image branch that makes predictions directly from the RGB image features. At the core of our method is a cross-modal transformer block that allows information to flow across these two branches by modeling the attention between 2D keypoint coordinates and image spatial features. Our architecture is smartly designed, which enables us to train on various types of datasets including images with 2D/3D annotations, images with 3D pseudo labels, and motion capture datasets that do not have associated images. This effectively improves the accuracy and generalization ability of our system. Built on a lightweight backbone (MobileNetV3), our method runs blazing fast (over 30fps on a single CPU core) and still yields competitive accuracy. Furthermore, with an HRNet backbone, XFormer delivers state-of-the-art performance on Huamn3.6 and 3DPW datasets.
Keyword: pruning
Less Can Be More: Unsupervised Graph Pruning for Large-scale Dynamic Graphs
Abstract
The prevalence of large-scale graphs poses great challenges in time and storage for training and deploying graph neural networks (GNNs). Several recent works have explored solutions for pruning the large original graph into a small and highly-informative one, such that training and inference on the pruned and large graphs have comparable performance. Although empirically effective, current researches focus on static or non-temporal graphs, which are not directly applicable to dynamic scenarios. In addition, they require labels as ground truth to learn the informative structure, limiting their applicability to new problem domains where labels are hard to obtain. To solve the dilemma, we propose and study the problem of unsupervised graph pruning on dynamic graphs. We approach the problem by our proposed STEP, a self-supervised temporal pruning framework that learns to remove potentially redundant edges from input dynamic graphs. From a technical and industrial viewpoint, our method overcomes the trade-offs between the performance and the time & memory overheads. Our results on three real-world datasets demonstrate the advantages on improving the efficacy, robustness, and efficiency of GNNs on dynamic node classification tasks. Most notably, STEP is able to prune more than 50% of edges on a million-scale industrial graph Alipay (7M nodes, 21M edges) while approximating up to 98% of the original performance. Code is available at https://github.com/EdisonLeeeee/STEP.
Boost Vision Transformer with GPU-Friendly Sparsity and Quantization
Authors: Chong Yu, Tao Chen, Zhongxue Gan, Jiayuan Fan
Abstract
The transformer extends its success from the language to the vision domain. Because of the stacked self-attention and cross-attention blocks, the acceleration deployment of vision transformer on GPU hardware is challenging and also rarely studied. This paper thoroughly designs a compression scheme to maximally utilize the GPU-friendly 2:4 fine-grained structured sparsity and quantization. Specially, an original large model with dense weight parameters is first pruned into a sparse one by 2:4 structured pruning, which considers the GPU's acceleration of 2:4 structured sparse pattern with FP16 data type, then the floating-point sparse model is further quantized into a fixed-point one by sparse-distillation-aware quantization aware training, which considers GPU can provide an extra speedup of 2:4 sparse calculation with integer tensors. A mixed-strategy knowledge distillation is used during the pruning and quantization process. The proposed compression scheme is flexible to support supervised and unsupervised learning styles. Experiment results show GPUSQ-ViT scheme achieves state-of-the-art compression by reducing vision transformer models 6.4-12.7 times on model size and 30.3-62 times on FLOPs with negligible accuracy degradation on ImageNet classification, COCO detection and ADE20K segmentation benchmarking tasks. Moreover, GPUSQ-ViT can boost actual deployment performance by 1.39-1.79 times and 3.22-3.43 times of latency and throughput on A100 GPU, and 1.57-1.69 times and 2.11-2.51 times improvement of latency and throughput on AGX Orin.
Abstract
Generative modeling has recently undergone remarkable advancements, primarily propelled by the transformative implications of Diffusion Probabilistic Models (DPMs). The impressive capability of these models, however, often entails significant computational overhead during both training and inference. To tackle this challenge, we present Diff-Pruning, an efficient compression method tailored for learning lightweight diffusion models from pre-existing ones, without the need for extensive re-training. The essence of Diff-Pruning is encapsulated in a Taylor expansion over pruned timesteps, a process that disregards non-contributory diffusion steps and ensembles informative gradients to identify important weights. Our empirical assessment, undertaken across four diverse datasets highlights two primary benefits of our proposed method: 1) Efficiency: it enables approximately a 50% reduction in FLOPs at a mere 10% to 20% of the original training expenditure; 2) Consistency: the pruned diffusion models inherently preserve generative behavior congruent with their pre-trained progenitors. Code is available at \url{https://github.com/VainF/Diff-Pruning}.
Learning Activation Functions for Sparse Neural Networks
Authors: Mohammad Loni, Aditya Mohan, Mehdi Asadi, Marius Lindauer
Subjects: Machine Learning (cs.LG); Neural and Evolutionary Computing (cs.NE)
Abstract
Sparse Neural Networks (SNNs) can potentially demonstrate similar performance to their dense counterparts while saving significant energy and memory at inference. However, the accuracy drop incurred by SNNs, especially at high pruning ratios, can be an issue in critical deployment conditions. While recent works mitigate this issue through sophisticated pruning techniques, we shift our focus to an overlooked factor: hyperparameters and activation functions. Our analyses have shown that the accuracy drop can additionally be attributed to (i) Using ReLU as the default choice for activation functions unanimously, and (ii) Fine-tuning SNNs with the same hyperparameters as dense counterparts. Thus, we focus on learning a novel way to tune activation functions for sparse networks and combining these with a separate hyperparameter optimization (HPO) regime for sparse networks. By conducting experiments on popular DNN models (LeNet-5, VGG-16, ResNet-18, and EfficientNet-B0) trained on MNIST, CIFAR-10, and ImageNet-16 datasets, we show that the novel combination of these two approaches, dubbed Sparse Activation Function Search, short: SAFS, results in up to 15.53%, 8.88%, and 6.33% absolute improvement in the accuracy for LeNet-5, VGG-16, and ResNet-18 over the default training protocols, especially at high pruning ratios. Our code can be found at https://github.com/automl/SAFS
Keyword: voxel
There is no result
Keyword: lidar
Improving Extrinsics between RADAR and LIDAR using Learning
Authors: Peng Jiang, Srikanth Saripalli
Subjects: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
Abstract
LIDAR and RADAR are two commonly used sensors in autonomous driving systems. The extrinsic calibration between the two is crucial for effective sensor fusion. The challenge arises due to the low accuracy and sparse information in RADAR measurements. This paper presents a novel solution for 3D RADAR-LIDAR calibration in autonomous systems. The method employs simple targets to generate data, including correspondence registration and a one-step optimization algorithm. The optimization aims to minimize the reprojection error while utilizing a small multi-layer perception (MLP) to perform regression on the return energy of the sensor around the targets. The proposed approach uses a deep learning framework such as PyTorch and can be optimized through gradient descent. The experiment uses a 360-degree Ouster-128 LIDAR and a 360-degree Navtech RADAR, providing raw measurements. The results validate the effectiveness of the proposed method in achieving improved estimates of extrinsic calibration parameters.
Keyword: diffusion
Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models
Abstract
Despite tremendous progress in generating high-quality images using diffusion models, synthesizing a sequence of animated frames that are both photorealistic and temporally coherent is still in its infancy. While off-the-shelf billion-scale datasets for image generation are available, collecting similar video data of the same scale is still challenging. Also, training a video diffusion model is computationally much more expensive than its image counterpart. In this work, we explore finetuning a pretrained image diffusion model with video data as a practical solution for the video synthesis task. We find that naively extending the image noise prior to video noise prior in video diffusion leads to sub-optimal performance. Our carefully designed video noise prior leads to substantially better performance. Extensive experimental validation shows that our model, Preserve Your Own Correlation (PYoCo), attains SOTA zero-shot text-to-video results on the UCF-101 and MSR-VTT benchmarks. It also achieves SOTA video generation quality on the small-scale UCF-101 benchmark with a $10\times$ smaller model using significantly less computation than the prior art.
PTQD: Accurate Post-Training Quantization for Diffusion Models
Abstract
Diffusion models have recently dominated image synthesis and other related generative tasks. However, the iterative denoising process is expensive in computations at inference time, making diffusion models less practical for low-latency and scalable real-world applications. Post-training quantization of diffusion models can significantly reduce the model size and accelerate the sampling process without requiring any re-training. Nonetheless, applying existing post-training quantization methods directly to low-bit diffusion models can significantly impair the quality of generated samples. Specifically, for each denoising step, quantization noise leads to deviations in the estimated mean and mismatches with the predetermined variance schedule. Moreover, as the sampling process proceeds, the quantization noise may accumulate, resulting in a low signal-to-noise ratio (SNR) in late denoising steps. To address these challenges, we propose a unified formulation for the quantization noise and diffusion perturbed noise in the quantized denoising process. We first disentangle the quantization noise into its correlated and residual uncorrelated parts regarding its full-precision counterpart. The correlated part can be easily corrected by estimating the correlation coefficient. For the uncorrelated part, we calibrate the denoising variance schedule to absorb the excess variance resulting from quantization. Moreover, we propose a mixed-precision scheme to choose the optimal bitwidth for each denoising step, which prefers low bits to accelerate the early denoising steps while high bits maintain the high SNR for the late steps. Extensive experiments demonstrate that our method outperforms previous post-training quantized diffusion models in generating high-quality samples, with only a 0.06 increase in FID score compared to full-precision LDM-4 on ImageNet 256x256, while saving 19.9x bit operations.
Abstract
Unrestricted adversarial attacks typically manipulate the semantic content of an image (e.g., color or texture) to create adversarial examples that are both effective and photorealistic, demonstrating their ability to deceive human perception and deep neural networks with stealth and success. However, current works usually sacrifice unrestricted degrees and subjectively select some image content to guarantee the photorealism of unrestricted adversarial examples, which limits its attack performance. To ensure the photorealism of adversarial examples and boost attack performance, we propose a novel unrestricted attack framework called Content-based Unrestricted Adversarial Attack. By leveraging a low-dimensional manifold that represents natural images, we map the images onto the manifold and optimize them along its adversarial direction. Therefore, within this framework, we implement Adversarial Content Attack based on Stable Diffusion and can generate high transferable unrestricted adversarial examples with various adversarial contents. Extensive experimentation and visualization demonstrate the efficacy of ACA, particularly in surpassing state-of-the-art attacks by an average of 13.3-50.4% and 16.8-48.0% in normally trained models and defense methods, respectively.
RMSSinger: Realistic-Music-Score based Singing Voice Synthesis
Abstract
We are interested in a challenging task, Realistic-Music-Score based Singing Voice Synthesis (RMS-SVS). RMS-SVS aims to generate high-quality singing voices given realistic music scores with different note types (grace, slur, rest, etc.). Though significant progress has been achieved, recent singing voice synthesis (SVS) methods are limited to fine-grained music scores, which require a complicated data collection pipeline with time-consuming manual annotation to align music notes with phonemes. Furthermore, these manual annotation destroys the regularity of note durations in music scores, making fine-grained music scores inconvenient for composing. To tackle these challenges, we propose RMSSinger, the first RMS-SVS method, which takes realistic music scores as input, eliminating most of the tedious manual annotation and avoiding the aforementioned inconvenience. Note that music scores are based on words rather than phonemes, in RMSSinger, we introduce word-level modeling to avoid the time-consuming phoneme duration annotation and the complicated phoneme-level mel-note alignment. Furthermore, we propose the first diffusion-based pitch modeling method, which ameliorates the naturalness of existing pitch-modeling methods. To achieve these, we collect a new dataset containing realistic music scores and singing voices according to these realistic music scores from professional singers. Extensive experiments on the dataset demonstrate the effectiveness of our methods. Audio samples are available at https://rmssinger.github.io/.
Abstract
Diffusions are a successful technique to sample from high-dimensional distributions can be either explicitly given or learnt from a collection of samples. They implement a diffusion process whose endpoint is a sample from the target distribution and whose drift is typically represented as a neural network. Stochastic localization is a successful technique to prove mixing of Markov Chains and other functional inequalities in high dimension. An algorithmic version of stochastic localization was introduced in [EAMS2022], to obtain an algorithm that samples from certain statistical mechanics models. This notes have three objectives: (i) Generalize the construction [EAMS2022] to other stochastic localization processes; (ii) Clarify the connection between diffusions and stochastic localization. In particular we show that standard denoising diffusions are stochastic localizations but other examples that are naturally suggested by the proposed viewpoint; (iii) Describe some insights that follow from this viewpoint.
Dirichlet Diffusion Score Model for Biological Sequence Generation
Abstract
Designing biological sequences is an important challenge that requires satisfying complex constraints and thus is a natural problem to address with deep generative modeling. Diffusion generative models have achieved considerable success in many applications. Score-based generative stochastic differential equations (SDE) model is a continuous-time diffusion model framework that enjoys many benefits, but the originally proposed SDEs are not naturally designed for modeling discrete data. To develop generative SDE models for discrete data such as biological sequences, here we introduce a diffusion process defined in the probability simplex space with stationary distribution being the Dirichlet distribution. This makes diffusion in continuous space natural for modeling discrete data. We refer to this approach as Dirchlet diffusion score model. We demonstrate that this technique can generate samples that satisfy hard constraints using a Sudoku generation task. This generative model can also solve Sudoku, including hard puzzles, without additional training. Finally, we applied this approach to develop the first human promoter DNA sequence design model and showed that designed sequences share similar properties with natural promoter sequences.
Zero-Day Backdoor Attack against Text-to-Image Diffusion Models via Personalization
Authors: Yihao Huang, Qing Guo, Felix Juefei-Xu
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Although recent personalization methods have democratized high-resolution image synthesis by enabling swift concept acquisition with minimal examples and lightweight computation, they also present an exploitable avenue for high accessible backdoor attacks. This paper investigates a critical and unexplored aspect of text-to-image (T2I) diffusion models - their potential vulnerability to backdoor attacks via personalization. Our study focuses on a zero-day backdoor vulnerability prevalent in two families of personalization methods, epitomized by Textual Inversion and DreamBooth.Compared to traditional backdoor attacks, our proposed method can facilitate more precise, efficient, and easily accessible attacks with a lower barrier to entry. We provide a comprehensive review of personalization in T2I diffusion models, highlighting the operation and exploitation potential of this backdoor vulnerability. To be specific, by studying the prompt processing of Textual Inversion and DreamBooth, we have devised dedicated backdoor attacks according to the different ways of dealing with unseen tokens and analyzed the influence of triggers and concept images on the attack effect. Our empirical study has shown that the nouveau-token backdoor attack has better attack performance while legacy-token backdoor attack is potentially harder to defend.
Discriminative Diffusion Models as Few-shot Vision and Language Learners
Authors: Xuehai He, Weixi Feng, Tsu-Jui Fu, Varun Jampani, Arjun Akula, Pradyumna Narayana, Sugato Basu, William Yang Wang, Xin Eric Wang
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Diffusion models, such as Stable Diffusion, have shown incredible performance on text-to-image generation. Since text-to-image generation often requires models to generate visual concepts with fine-grained details and attributes specified in text prompts, can we leverage the powerful representations learned by pre-trained diffusion models for discriminative tasks such as image-text matching? To answer this question, we propose a novel approach, Discriminative Stable Diffusion (DSD), which turns pre-trained text-to-image diffusion models into few-shot discriminative learners. Our approach uses the cross-attention score of a Stable Diffusion model to capture the mutual influence between visual and textual information and fine-tune the model via attention-based prompt learning to perform image-text matching. By comparing DSD with state-of-the-art methods on several benchmark datasets, we demonstrate the potential of using pre-trained diffusion models for discriminative tasks with superior results on few-shot image-text matching.
Diffusion-Based Speech Enhancement with Joint Generative and Predictive Decoders
Abstract
Diffusion-based speech enhancement (SE) has been investigated recently, but its decoding is very time-consuming. One solution is to initialize the decoding process with the enhanced feature estimated by a predictive SE system. However, this two-stage method ignores the complementarity between predictive and diffusion SE. In this paper, we propose a unified system that integrates these two SE modules. The system encodes both generative and predictive information, and then applies both generative and predictive decoders, whose outputs are fused. Specifically, the two SE modules are fused in the first and final diffusion steps: the first step fusion initializes the diffusion process with the predictive SE for improving the convergence, and the final step fusion combines the two complementary SE outputs to improve the SE performance. Experiments on the Voice-Bank dataset show that the diffusion score estimation can benefit from the predictive information and speed up the decoding.
Multi-resolution Spatiotemporal Enhanced Transformer Denoising with Functional Diffusive GANs for Constructing Brain Effective Connectivity in MCI analysis
Authors: Qiankun Zuo, Chi-Man Pun, Yudong Zhang, Hongfei Wang, Jin Hong
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
Effective connectivity can describe the causal patterns among brain regions. These patterns have the potential to reveal the pathological mechanism and promote early diagnosis and effective drug development for cognitive disease. However, the current studies mainly focus on using empirical functional time series to calculate effective connections, which may not comprehensively capture the complex causal relationships between brain regions. In this paper, a novel Multi-resolution Spatiotemporal Enhanced Transformer Denoising (MSETD) network with an adversarially functional diffusion model is proposed to map functional magnetic resonance imaging (fMRI) into effective connectivity for mild cognitive impairment (MCI) analysis. To be specific, the denoising framework leverages a conditional diffusion process that progressively translates the noise and conditioning fMRI to effective connectivity in an end-to-end manner. To ensure reverse diffusion quality and diversity, the multi-resolution enhanced transformer generator is designed to extract local and global spatiotemporal features. Furthermore, a multi-scale diffusive transformer discriminator is devised to capture the temporal patterns at different scales for generation stability. Evaluations of the ADNI datasets demonstrate the feasibility and efficacy of the proposed model. The proposed model not only achieves superior prediction performance compared with other competing methods but also identifies MCI-related causal connections that are consistent with clinical studies.
Catch-Up Distillation: You Only Need to Train Once for Accelerating Sampling
Authors: Shitong Shao, Xu Dai, Shouyi Yin, Lujun Li, Huanran Chen, Yang Hu
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Diffusion Probability Models (DPMs) have made impressive advancements in various machine learning domains. However, achieving high-quality synthetic samples typically involves performing a large number of sampling steps, which impedes the possibility of real-time sample synthesis. Traditional accelerated sampling algorithms via knowledge distillation rely on pre-trained model weights and discrete time step scenarios, necessitating additional training sessions to achieve their goals. To address these issues, we propose the Catch-Up Distillation (CUD), which encourages the current moment output of the velocity estimation model ``catch up'' with its previous moment output. Specifically, CUD adjusts the original Ordinary Differential Equation (ODE) training objective to align the current moment output with both the ground truth label and the previous moment output, utilizing Runge-Kutta-based multi-step alignment distillation for precise ODE estimation while preventing asynchronous updates. Furthermore, we investigate the design space for CUDs under continuous time-step scenarios and analyze how to determine the suitable strategies. To demonstrate CUD's effectiveness, we conduct thorough ablation and comparison experiments on CIFAR-10, MNIST, and ImageNet-64. On CIFAR-10, we obtain a FID of 2.80 by sampling in 15 steps under one-session training and the new state-of-the-art FID of 3.37 by sampling in one step with additional training. This latter result necessitated only 62w iterations with a batch size of 128, in contrast to Consistency Distillation, which demanded 210w iterations with a larger batch size of 256.
Supercloseness of the LDG method for a two-dimensional singularly perturbed convection-diffusion problem on Bakhvalov-type mesh
Abstract
In this paper, we focus on analyzing the supercloseness property of a two-dimensional singularly perturbed convection-diffusion problem with exponential boundary layers. The local discontinuous Galerkin (LDG) method with piecewise tensor-product polynomials of degree k is applied to Bakhvalov-type mesh. By developing special two-dimensional local Gauss-Radau projections and establishing a novel interpolation, supercloseness of an optimal order k+1 can be achieved on Bakhvalov-type mesh. It is crucial to highlight that this supercloseness result is independent of the singular perturbation parameter.
Democratized Diffusion Language Model
Authors: Nikita Balagansky, Daniil Gavrilov
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Abstract
Despite the potential benefits of Diffusion Models for NLP applications, publicly available implementations, trained models, or reproducible training procedures currently need to be publicly available. We present the Democratized Diffusion Language Model (DDLM), based on the Continuous Diffusion for Categorical Data (CDCD) framework, to address these challenges. We propose a simplified training procedure for DDLM using the C4 dataset and perform an in-depth analysis of the trained model's behavior. Furthermore, we introduce a novel early-exiting strategy for faster sampling with models trained with score interpolation. Since no previous works aimed at solving downstream tasks with pre-trained Diffusion LM (e.g., classification tasks), we experimented with GLUE Benchmark to study the ability of DDLM to transfer knowledge. With this paper, we propose available training and evaluation pipelines to other researchers and pre-trained DDLM models, which could be used in future research with Diffusion LMs.
Abstract
Diffusion model based language-guided image editing has achieved great success recently. However, existing state-of-the-art diffusion models struggle with rendering correct text and text style during generation. To tackle this problem, we propose a universal self-supervised text editing diffusion model (DiffUTE), which aims to replace or modify words in the source image with another one while maintaining its realistic appearance. Specifically, we build our model on a diffusion model and carefully modify the network structure to enable the model for drawing multilingual characters with the help of glyph and position information. Moreover, we design a self-supervised learning framework to leverage large amounts of web data to improve the representation ability of the model. Experimental results show that our method achieves an impressive performance and enables controllable editing on in-the-wild images with high fidelity. Our code will be avaliable in \url{https://github.com/chenhaoxing/DiffUTE}.
Constructing a personalized AI assistant for shear wall layout using Stable Diffusion
Abstract
Shear wall structures are widely used in high-rise residential buildings, and the layout of shear walls requires many years of design experience and iterative trial and error. Currently, there are methods based on heuristic algorithms, but they generate results too slowly. Those based on Generative Adversarial Networks (GANs) or Graph Neural Networks (GNNs) can only generate single arrangements and require large amounts of training data. At present, Stable Diffusion is being widely used, and by using the Low-Rank Adaptation (LoRA) method to fine-tune large models with small amounts of data, good generative results can be achieved. Therefore, this paper proposes a personalized AI assistant for shear wall layout based on Stable Diffusion, which has been proven to produce good generative results through testing.
AIwriting: Relations Between Image Generation and Digital Writing
Authors: Scott Rettberg, Talan Memmott, Jill Walker Rettberg, Jason Nelson, Patrick Lichty
Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Human-Computer Interaction (cs.HC); Multimedia (cs.MM)
Abstract
During 2022, both transformer-based AI text generation sys-tems such as GPT-3 and AI text-to-image generation systems such as DALL-E 2 and Stable Diffusion made exponential leaps forward and are unquestionably altering the fields of digital art and electronic literature. In this panel a group of electronic literature authors and theorists consider new oppor-tunities for human creativity presented by these systems and present new works have produced during the past year that specifically address these systems as environments for literary expressions that are translated through iterative interlocutive processes into visual representations. The premise that binds these presentations is that these systems and the works gener-ated must be considered from a literary perspective, as they originate in human writing. In works ranging from a visual memoir of the personal experience of a health crisis, to interac-tive web comics, to architectures based on abstract poetic language, to political satire, four artists explore the capabili-ties of these writing environments for new genres of literary artist practice, while a digital culture theorist considers the origins and effects of the particular training datasets of human language and images on which these new hybrid forms are based.
GETMusic: Generating Any Music Tracks with a Unified Representation and Diffusion Framework
Authors: Ang Lv, Xu Tan, Peiling Lu, Wei Ye, Shikun Zhang, Jiang Bian, Rui Yan
Abstract
Symbolic music generation aims to create musical notes, which can help users compose music, such as generating target instrumental tracks from scratch, or based on user-provided source tracks. Considering the diverse and flexible combination between source and target tracks, a unified model capable of generating any arbitrary tracks is of crucial necessity. Previous works fail to address this need due to inherent constraints in music representations and model architectures. To address this need, we propose a unified representation and diffusion framework named GETMusic (`GET' stands for GEnerate music Tracks), which includes a novel music representation named GETScore, and a diffusion model named GETDiff. GETScore represents notes as tokens and organizes them in a 2D structure, with tracks stacked vertically and progressing horizontally over time. During training, tracks are randomly selected as either the target or source. In the forward process, target tracks are corrupted by masking their tokens, while source tracks remain as ground truth. In the denoising process, GETDiff learns to predict the masked target tokens, conditioning on the source tracks. With separate tracks in GETScore and the non-autoregressive behavior of the model, GETMusic can explicitly control the generation of any target tracks from scratch or conditioning on source tracks. We conduct experiments on music generation involving six instrumental tracks, resulting in a total of 665 combinations. GETMusic provides high-quality results across diverse combinations and surpasses prior works proposed for some specific combinations.
X-IQE: eXplainable Image Quality Evaluation for Text-to-Image Generation with Visual Large Language Models
Authors: Yixiong Chen
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
This paper introduces a novel explainable image quality evaluation approach called X-IQE, which leverages visual large language models (LLMs) to evaluate text-to-image generation methods by generating textual explanations. X-IQE utilizes a hierarchical Chain of Thought (CoT) to enable MiniGPT-4 to produce self-consistent, unbiased texts that are highly correlated with human evaluation. It offers several advantages, including the ability to distinguish between real and generated images, evaluate text-image alignment, and assess image aesthetics without requiring model training or fine-tuning. X-IQE is more cost-effective and efficient compared to human evaluation, while significantly enhancing the transparency and explainability of deep image quality evaluation models. We validate the effectiveness of our method as a benchmark using images generated by prevalent diffusion models. X-IQE demonstrates similar performance to state-of-the-art (SOTA) evaluation methods on COCO Caption, while overcoming the limitations of previous evaluation models on DrawBench, particularly in handling ambiguous generation prompts and text recognition in generated images. Project website: https://github.com/Schuture/Benchmarking-Awesome-Diffusion-Models
LDM3D: Latent Diffusion Model for 3D
Authors: Gabriela Ben Melech Stan, Diana Wofk, Scottie Fox, Alex Redden, Will Saxton, Jean Yu, Estelle Aflalo, Shao-Yen Tseng, Fabio Nonato, Matthias Muller, Vasudev Lal
Subjects: Computer Vision and Pattern Recognition (cs.CV)
Abstract
This research paper proposes a Latent Diffusion Model for 3D (LDM3D) that generates both image and depth map data from a given text prompt, allowing users to generate RGBD images from text prompts. The LDM3D model is fine-tuned on a dataset of tuples containing an RGB image, depth map and caption, and validated through extensive experiments. We also develop an application called DepthFusion, which uses the generated RGB images and depth maps to create immersive and interactive 360-degree-view experiences using TouchDesigner. This technology has the potential to transform a wide range of industries, from entertainment and gaming to architecture and design. Overall, this paper presents a significant contribution to the field of generative AI and computer vision, and showcases the potential of LDM3D and DepthFusion to revolutionize content creation and digital experiences. A short video summarizing the approach can be found at https://t.ly/tdi2.
Abstract
Diffusion models have gained increasing attention for their impressive generation abilities but currently struggle with rendering accurate and coherent text. To address this issue, we introduce \textbf{TextDiffuser}, focusing on generating images with visually appealing text that is coherent with backgrounds. TextDiffuser consists of two stages: first, a Transformer model generates the layout of keywords extracted from text prompts, and then diffusion models generate images conditioned on the text prompt and the generated layout. Additionally, we contribute the first large-scale text images dataset with OCR annotations, \textbf{MARIO-10M}, containing 10 million image-text pairs with text recognition, detection, and character-level segmentation annotations. We further collect the \textbf{MARIO-Eval} benchmark to serve as a comprehensive tool for evaluating text rendering quality. Through experiments and user studies, we show that TextDiffuser is flexible and controllable to create high-quality text images using text prompts alone or together with text template images, and conduct text inpainting to reconstruct incomplete images with text. The code, model, and dataset will be available at \url{https://aka.ms/textdiffuser}.
Abstract
Generative modeling has recently undergone remarkable advancements, primarily propelled by the transformative implications of Diffusion Probabilistic Models (DPMs). The impressive capability of these models, however, often entails significant computational overhead during both training and inference. To tackle this challenge, we present Diff-Pruning, an efficient compression method tailored for learning lightweight diffusion models from pre-existing ones, without the need for extensive re-training. The essence of Diff-Pruning is encapsulated in a Taylor expansion over pruned timesteps, a process that disregards non-contributory diffusion steps and ensembles informative gradients to identify important weights. Our empirical assessment, undertaken across four diverse datasets highlights two primary benefits of our proposed method: 1) Efficiency: it enables approximately a 50% reduction in FLOPs at a mere 10% to 20% of the original training expenditure; 2) Consistency: the pruned diffusion models inherently preserve generative behavior congruent with their pre-trained progenitors. Code is available at \url{https://github.com/VainF/Diff-Pruning}.
Unsupervised Pansharpening via Low-rank Diffusion Model
Abstract
Pansharpening is a process of merging a highresolution panchromatic (PAN) image and a low-resolution multispectral (LRMS) image to create a single high-resolution multispectral (HRMS) image. Most of the existing deep learningbased pansharpening methods have poor generalization ability and the traditional model-based pansharpening methods need careful manual exploration for the image structure prior. To alleviate these issues, this paper proposes an unsupervised pansharpening method by combining the diffusion model with the low-rank matrix factorization technique. Specifically, we assume that the HRMS image is decomposed into the product of two low-rank tensors, i.e., the base tensor and the coefficient matrix. The base tensor lies on the image field and has low spectral dimension, we can thus conveniently utilize a pre-trained remote sensing diffusion model to capture its image structures. Additionally, we derive a simple yet quite effective way to preestimate the coefficient matrix from the observed LRMS image, which preserves the spectral information of the HRMS. Extensive experimental results on some benchmark datasets demonstrate that our proposed method performs better than traditional model-based approaches and has better generalization ability than deep learning-based techniques. The code is released in https://github.com/xyrui/PLRDiff.
Generating coherent comic with rich story using ChatGPT and Stable Diffusion
Abstract
Past work demonstrated that using neural networks, we can extend unfinished music pieces while maintaining the music style of the musician. With recent advancements in large language models and diffusion models, we are now capable of generating comics with an interesting storyline while maintaining the art style of the artist. In this paper, we used ChatGPT to generate storylines and dialogue and then generated the comic using stable diffusion. We introduced a novel way to evaluate AI-generated stories, and we achieved SOTA performance on character fidelity and art style by fine-tuning stable diffusion using LoRA, ControlNet, etc.
Inspecting the Geographical Representativeness of Images from Text-to-Image Models
Authors: Abhipsa Basu, R. Venkatesh Babu, Danish Pruthi
Subjects: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
Abstract
Recent progress in generative models has resulted in models that produce both realistic as well as relevant images for most textual inputs. These models are being used to generate millions of images everyday, and hold the potential to drastically impact areas such as generative art, digital marketing and data augmentation. Given their outsized impact, it is important to ensure that the generated content reflects the artifacts and surroundings across the globe, rather than over-representing certain parts of the world. In this paper, we measure the geographical representativeness of common nouns (e.g., a house) generated through DALL.E 2 and Stable Diffusion models using a crowdsourced study comprising 540 participants across 27 countries. For deliberately underspecified inputs without country names, the generated images most reflect the surroundings of the United States followed by India, and the top generations rarely reflect surroundings from all other countries (average score less than 3 out of 5). Specifying the country names in the input increases the representativeness by 1.44 points on average for DALL.E 2 and 0.75 for Stable Diffusion, however, the overall scores for many countries still remain low, highlighting the need for future models to be more geographically inclusive. Lastly, we examine the feasibility of quantifying the geographical representativeness of generated images without conducting user studies.
Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces
Authors: Javier E Santos, Zachary R. Fox, Nicholas Lubbers, Yen Ting Lin
Subjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Typical generative diffusion models rely on a Gaussian diffusion process for training the backward transformations, which can then be used to generate samples from Gaussian noise. However, real world data often takes place in discrete-state spaces, including many scientific applications. Here, we develop a theoretical formulation for arbitrary discrete-state Markov processes in the forward diffusion process using exact (as opposed to variational) analysis. We relate the theory to the existing continuous-state Gaussian diffusion as well as other approaches to discrete diffusion, and identify the corresponding reverse-time stochastic process and score function in the continuous-time setting, and the reverse-time mapping in the discrete-time setting. As an example of this framework, we introduce ``Blackout Diffusion'', which learns to produce samples from an empty image instead of from noise. Numerical experiments on the CIFAR-10, Binarized MNIST, and CelebA datasets confirm the feasibility of our approach. Generalizing from specific (Gaussian) forward processes to discrete-state processes without a variational approximation sheds light on how to interpret diffusion models, which we discuss.
UniControl: A Unified Diffusion Model for Controllable Visual Generation In the Wild
Authors: Can Qin, Shu Zhang, Ning Yu, Yihao Feng, Xinyi Yang, Yingbo Zhou, Huan Wang, Juan Carlos Niebles, Caiming Xiong, Silvio Savarese, Stefano Ermon, Yun Fu, Ran Xu
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Abstract
Achieving machine autonomy and human control often represent divergent objectives in the design of interactive AI systems. Visual generative foundation models such as Stable Diffusion show promise in navigating these goals, especially when prompted with arbitrary languages. However, they often fall short in generating images with spatial, structural, or geometric controls. The integration of such controls, which can accommodate various visual conditions in a single unified model, remains an unaddressed challenge. In response, we introduce UniControl, a new generative foundation model that consolidates a wide array of controllable condition-to-image (C2I) tasks within a singular framework, while still allowing for arbitrary language prompts. UniControl enables pixel-level-precise image generation, where visual conditions primarily influence the generated structures and language prompts guide the style and context. To equip UniControl with the capacity to handle diverse visual conditions, we augment pretrained text-to-image diffusion models and introduce a task-aware HyperNet to modulate the diffusion models, enabling the adaptation to different C2I tasks simultaneously. Trained on nine unique C2I tasks, UniControl demonstrates impressive zero-shot generation abilities with unseen visual conditions. Experimental results show that UniControl often surpasses the performance of single-task-controlled methods of comparable model sizes. This control versatility positions UniControl as a significant advancement in the realm of controllable visual generation.
Keyword: dynamic
Bringing AI to the edge: A formal M&S specification to deploy effective IoT architectures
Authors: Román Cárdenas, Patricia Arroba, José L. Risco-Martín
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
The Internet of Things is transforming our society, providing new services that improve the quality of life and resource management. These applications are based on ubiquitous networks of multiple distributed devices, with limited computing resources and power, capable of collecting and storing data from heterogeneous sources in real-time. To avoid network saturation and high delays, new architectures such as fog computing are emerging to bring computing infrastructure closer to data sources. Additionally, new data centers are needed to provide real-time Big Data and data analytics capabilities at the edge of the network, where energy efficiency needs to be considered to ensure a sustainable and effective deployment in areas of human activity. In this research, we present an IoT model based on the principles of Model-Based Systems Engineering defined using the Discrete Event System Specification formalism. The provided mathematical formalism covers the description of the entire architecture, from IoT devices to the processing units in edge data centers. Our work includes the location-awareness of user equipment, network, and computing infrastructures to optimize federated resource management in terms of delay and power consumption. We present an effective framework to assist the dimensioning and the dynamic operation of IoT data stream analytics applications, demonstrating our contributions through a driving assistance use case based on real traces and data.
Intelligent multicast routing method based on multi-agent deep reinforcement learning in SDWN
Authors: Hongwen Hu, Miao Ye, Chenwei Zhao, Qiuxiang Jiang, Yong Wang, Hongbing Qiu, Xiaofang Deng
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
Multicast communication technology is widely applied in wireless environments with a high device density. Traditional wireless network architectures have difficulty flexibly obtaining and maintaining global network state information and cannot quickly respond to network state changes, thus affecting the throughput, delay, and other QoS requirements of existing multicasting solutions. Therefore, this paper proposes a new multicast routing method based on multiagent deep reinforcement learning (MADRL-MR) in a software-defined wireless networking (SDWN) environment. First, SDWN technology is adopted to flexibly configure the network and obtain network state information in the form of traffic matrices representing global network links information, such as link bandwidth, delay, and packet loss rate. Second, the multicast routing problem is divided into multiple subproblems, which are solved through multiagent cooperation. To enable each agent to accurately understand the current network state and the status of multicast tree construction, the state space of each agent is designed based on the traffic and multicast tree status matrices, and the set of AP nodes in the network is used as the action space. A novel single-hop action strategy is designed, along with a reward function based on the four states that may occur during tree construction: progress, invalid, loop, and termination. Finally, a decentralized training approach is combined with transfer learning to enable each agent to quickly adapt to dynamic network changes and accelerate convergence. Simulation experiments show that MADRL-MR outperforms existing algorithms in terms of throughput, delay, packet loss rate, etc., and can establish more intelligent multicast routes.
An Intelligent SDWN Routing Algorithm Based on Network Situational Awareness and Deep Reinforcement Learning
Authors: Jinqiang Li, Miao Ye, Linqiang Huang, Xiaofang Deng, Hongbing Qiu, Yong Wang
Subjects: Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Abstract
Due to the highly dynamic changes in wireless network topologies, efficiently obtaining network status information and flexibly forwarding data to improve communication quality of service are important challenges. This article introduces an intelligent routing algorithm (DRL-PPONSA) based on proximal policy optimization deep reinforcement learning with network situational awareness under a software-defined wireless networking architecture. First, a specific data plane is designed for network topology construction and data forwarding. The control plane collects network traffic information, sends flow tables, and uses a GCN-GRU prediction mechanism to perceive future traffic change trends to achieve network situational awareness. Second, a DRL-based data forwarding mechanism is designed in the knowledge plane. The predicted network traffic matrix and topology information matrix are treated as the environment for DRL agents, while next-hop adjacent nodes are treated as executable actions. Accordingly, action selection strategies are designed for different network conditions to achieve more intelligent, flexible, and efficient routing control. The reward function is designed using network link information and various reward and penalty mechanisms. Additionally, importance sampling and gradient clipping techniques are employed during gradient updating to enhance convergence speed and stability. Experimental results show that DRL-PPONSA outperforms traditional routing methods in network throughput, delay, packet loss rate, and wireless node distance. Compared to value-function-based Dueling DQN routing, the convergence speed is significantly improved, and the convergence effect is more stable. Simultaneously, its consumption of hardware storage space is reduced, and efficient routing decisions can be made in real-time using the current network state information.
Emotion Recognition based on Psychological Components in Guided Narratives for Emotion Regulation
Authors: Gustave Cortal (LMF, LISN), Alain Finkel (LMF, IUF), Patrick Paroubek (LISN), Lina Ye (LMF)
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
Emotion regulation is a crucial element in dealing with emotional events and has positive effects on mental health. This paper aims to provide a more comprehensive understanding of emotional events by introducing a new French corpus of emotional narratives collected using a questionnaire for emotion regulation. We follow the theoretical framework of the Component Process Model which considers emotions as dynamic processes composed of four interrelated components (behavior, feeling, thinking and territory). Each narrative is related to a discrete emotion and is structured based on all emotion components by the writers. We study the interaction of components and their impact on emotion classification with machine learning methods and pre-trained language models. Our results show that each component improves prediction performance, and that the best results are achieved by jointly considering all components. Our results also show the effectiveness of pre-trained language models in predicting discrete emotion from certain components, which reveal differences in how emotion components are expressed.
The Effectiveness of a Dynamic Loss Function in Neural Network Based Automated Essay Scoring
Authors: Oscar Morris
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
Neural networks and in particular the attention mechanism have brought significant advances to the field of Automated Essay Scoring. Many of these systems use a regression-based model which may be prone to underfitting when the model only predicts the mean of the training data. In this paper, we present a dynamic loss function that creates an incentive for the model to predict with the correct distribution, as well as predicting the correct values. Our loss function achieves this goal without sacrificing any performance achieving a Quadratic Weighted Kappa score of 0.752 on the Automated Student Assessment Prize Automated Essay Scoring dataset.
Exact Recovery for System Identification with More Corrupt Data than Clean Data
Authors: Baturalp Yalcin, Javad Lavaei, Murat Arcak
Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC)
Abstract
In this paper, we study the system identification problem for linear discrete-time systems under adversaries and analyze two lasso-type estimators. We study both asymptotic and non-asymptotic properties of these estimators in two separate scenarios, corresponding to deterministic and stochastic models for the attack times. Since the samples collected from the system are correlated, the existing results on lasso are not applicable. We show that when the system is stable and the attacks are injected periodically, the sample complexity for the exact recovery of the system dynamics is O(n), where n is the dimension of the states. When the adversarial attacks occur at each time instance with probability p, the required sample complexity for the exact recovery scales as O(\log(n)p/(1-p)^2). This result implies the almost sure convergence to the true system dynamics under the asymptotic regime. As a by-product, even when more than half of the data is compromised, our estimators still learn the system correctly. This paper provides the first mathematical guarantee in the literature on learning from correlated data for dynamical systems in the case when there is less clean data than corrupt data.
Analytic relationship of relative synchronizability to network structure and motifs
Authors: Joseph T. Lizier, Frank Bauer, Fatihcan M. Atay, Jürgen Jost
Subjects: Social and Information Networks (cs.SI); Disordered Systems and Neural Networks (cond-mat.dis-nn); Mathematical Physics (math-ph); Physics and Society (physics.soc-ph)
Abstract
Synchronization phenomena on networks have attracted much attention in studies of neural, social, economic, and biological systems, yet we still lack a systematic understanding of how relative synchronizability relates to underlying network structure. Indeed, this question is of central importance to the key theme of how dynamics on networks relate to their structure more generally. We present an analytic technique to directly measure the relative synchronizability of noise-driven time-series processes on networks, in terms of the directed network structure. We consider both discrete-time auto-regressive processes and continuous-time Ornstein-Uhlenbeck dynamics on networks. Our technique builds on computation of the network covariance matrix in the space orthogonal to the synchronized state, enabling it to be more general than previous work in not requiring either symmetric (undirected) or diagonalizable connectivity matrices, and allowing arbitrary self-link weights. More importantly, our approach quantifies the relative synchronisation specifically in terms of the contribution of process motif (walk) structures. We demonstrate that in general the relative abundance of process motifs with convergent directed walks (including feedback and feedforward loops) hinders synchronizability. We also reveal subtle differences between the motifs involved for discrete or continuous-time dynamics. Our insights analytically explain several known general results regarding synchronizability of networks, including that small-world and regular networks are less synchronizable than random networks.
On a Doubly Reduced Model for Dynamics of Heterogeneous Mixtures of Stiffened Gases, its Regularizations and their Implementations
Abstract
We deal with the reduced four-equation model for dynamics of the heterogeneous compressible binary mixtures with the stiffened gas equations of state. We study its further reduced form, with the excluded volume concentrations and a quadratic equation for the common pressure of the components, that can be called quasi-homogeneous form. We prove new properties of this equation, derive a simple formula for the squared speed of sound, give an alternative proof for a formula that relates it to the squared Wood speed of sound, and a short derivation of the pressure balance equation. For the first time, we introduce regularizations of the heterogeneous model (in the quasi-homogeneous form). In the 1D case, we construct the corresponding explicit two-level in time and symmetric three-point in space finite-difference schemes without limiters and present various numerical results for flows with shock waves.
RAMP: Hierarchical Reactive Motion Planning for Manipulation Tasks Using Implicit Signed Distance Functions
Abstract
We introduce Reactive Action and Motion Planner (RAMP), which combines the strengths of search-based and reactive approaches for motion planning. In essence, RAMP is a hierarchical approach where a novel variant of a Model Predictive Path Integral (MPPI) controller is used to generate trajectories which are then followed asynchronously by a local vector field controller. We demonstrate, in the context of a table clearing application, that RAMP can rapidly find paths in the robot's configuration space, satisfy task and robot-specific constraints, and provide safety by reacting to static or dynamically moving obstacles. RAMP achieves superior performance through a number of key innovations: we use Signed Distance Function (SDF) representations directly from the robot configuration space, both for collision checking and reactive control. The use of SDFs allows for a smoother definition of collision cost when planning for a trajectory, and is critical in ensuring safety while following trajectories. In addition, we introduce a novel variant of MPPI which, combined with the safety guarantees of the vector field trajectory follower, performs incremental real-time global trajectory planning. Simulation results establish that our method can generate paths that are comparable to traditional and state-of-the-art approaches in terms of total trajectory length while being up to 30 times faster. Real-world experiments demonstrate the safety and effectiveness of our approach in challenging table clearing scenarios.
Discovering Individual Rewards in Collective Behavior through Inverse Multi-Agent Reinforcement Learning
Authors: Daniel Waelchli, Pascal Weber, Petros Koumoutsakos
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
Abstract
The discovery of individual objectives in collective behavior of complex dynamical systems such as fish schools and bacteria colonies is a long-standing challenge. Inverse reinforcement learning is a potent approach for addressing this challenge but its applicability to dynamical systems, involving continuous state-action spaces and multiple interacting agents, has been limited. In this study, we tackle this challenge by introducing an off-policy inverse multi-agent reinforcement learning algorithm (IMARL). Our approach combines the ReF-ER techniques with guided cost learning. By leveraging demonstrations, our algorithm automatically uncovers the reward function and learns an effective policy for the agents. Through extensive experimentation, we demonstrate that the proposed policy captures the behavior observed in the provided data, and achieves promising results across problem domains including single agent models in the OpenAI gym and multi-agent models of schooling behavior. The present study shows that the proposed IMARL algorithm is a significant step towards understanding collective dynamics from the perspective of its constituents, and showcases its value as a tool for studying complex physical systems exhibiting collective behaviour.
Sim-MEES: Modular End-Effector System Grasping Dataset for Mobile Manipulators in Cluttered Environments
Abstract
In this paper, we present Sim-MEES: a large-scale synthetic dataset that contains 1,550 objects with varying difficulty levels and physics properties, as well as 11 million grasp labels for mobile manipulators to plan grasps using different gripper modalities in cluttered environments. Our dataset generation process combines analytic models and dynamic simulations of the entire cluttered environment to provide accurate grasp labels. We provide a detailed study of our proposed labeling process for both parallel jaw grippers and suction cup grippers, comparing them with state-of-the-art methods to demonstrate how Sim-MEES can provide precise grasp labels in cluttered environments.
ACRoBat: Optimizing Auto-batching of Dynamic Deep Learning at Compile Time
Authors: Pratik Fegade, Tianqi Chen, Phillip B. Gibbons, Todd C. Mowry
Abstract
Dynamic control flow is an important technique often used to design expressive and efficient deep learning computations for applications such as text parsing, machine translation, exiting early out of deep models and so on. However, the resulting control flow divergence makes batching, an important performance optimization, difficult to perform manually. In this paper, we present ACRoBat, a framework that enables efficient automatic batching for dynamic deep learning computations by performing hybrid static+dynamic compiler optimizations and end-to-end tensor code generation. ACRoBat performs up to 8.5X better than DyNet, a state-of-the-art framework for automatic batching, on an Nvidia GeForce RTX 3070 GPU.
Learning Differentially Private Probabilistic Models for Privacy-Preserving Image Generation
Abstract
A number of deep models trained on high-quality and valuable images have been deployed in practical applications, which may pose a leakage risk of data privacy. Learning differentially private generative models can sidestep this challenge through indirect data access. However, such differentially private generative models learned by existing approaches can only generate images with a low-resolution of less than 128x128, hindering the widespread usage of generated images in downstream training. In this work, we propose learning differentially private probabilistic models (DPPM) to generate high-resolution images with differential privacy guarantee. In particular, we first train a model to fit the distribution of the training data and make it satisfy differential privacy by performing a randomized response mechanism during training process. Then we perform Hamiltonian dynamics sampling along with the differentially private movement direction predicted by the trained probabilistic model to obtain the privacy-preserving images. In this way, it is possible to apply these images to different downstream tasks while protecting private information. Notably, compared to other state-of-the-art differentially private generative approaches, our approach can generate images up to 256x256 with remarkable visual quality and data utility. Extensive experiments show the effectiveness of our approach.
Abstract
Relay Mining presents a scalable solution employing probabilistic mechanisms and crypto-economic incentives to estimate RPC volume usage, facilitating decentralized multi-tenant rate limiting. Network traffic from individual applications can be concurrently serviced by multiple RPC service providers, with costs, rewards, and rate limiting governed by a native cryptocurrency on a distributed ledger. Building upon established research in token bucket algorithms and distributed rate-limiting penalty models, our approach harnesses a feedback loop control mechanism to adjust the difficulty of mining relay rewards, dynamically scaling with network usage growth. By leveraging crypto-economic incentives, we reduce coordination overhead costs and introduce a mechanism for providing RPC services that are both geopolitically and geographically distributed.
Less Can Be More: Unsupervised Graph Pruning for Large-scale Dynamic Graphs
Abstract
The prevalence of large-scale graphs poses great challenges in time and storage for training and deploying graph neural networks (GNNs). Several recent works have explored solutions for pruning the large original graph into a small and highly-informative one, such that training and inference on the pruned and large graphs have comparable performance. Although empirically effective, current researches focus on static or non-temporal graphs, which are not directly applicable to dynamic scenarios. In addition, they require labels as ground truth to learn the informative structure, limiting their applicability to new problem domains where labels are hard to obtain. To solve the dilemma, we propose and study the problem of unsupervised graph pruning on dynamic graphs. We approach the problem by our proposed STEP, a self-supervised temporal pruning framework that learns to remove potentially redundant edges from input dynamic graphs. From a technical and industrial viewpoint, our method overcomes the trade-offs between the performance and the time & memory overheads. Our results on three real-world datasets demonstrate the advantages on improving the efficacy, robustness, and efficiency of GNNs on dynamic node classification tasks. Most notably, STEP is able to prune more than 50% of edges on a million-scale industrial graph Alipay (7M nodes, 21M edges) while approximating up to 98% of the original performance. Code is available at https://github.com/EdisonLeeeee/STEP.
Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learning
Authors: Yinglun Xu, Gagandeep Singh
Subjects: Machine Learning (cs.LG); Cryptography and Security (cs.CR)
Abstract
We propose the first black-box targeted attack against online deep reinforcement learning through reward poisoning during training time. Our attack is applicable to general environments with unknown dynamics learned by unknown algorithms and requires limited attack budgets and computational resources. We leverage a general framework and find conditions to ensure efficient attack under a general assumption of the learning algorithms. We show that our attack is optimal in our framework under the conditions. We experimentally verify that with limited budgets, our attack efficiently leads the learning agent to various target policies under a diverse set of popular DRL environments and state-of-the-art learners.
Paxion: Patching Action Knowledge in Video-Language Foundation Models
Abstract
Action knowledge involves the understanding of textual, visual, and temporal aspects of actions. We introduce the Action Dynamics Benchmark (ActionBench) containing two carefully designed probing tasks: Action Antonym and Video Reversal, which targets multimodal alignment capabilities and temporal understanding skills of the model, respectively. Despite recent video-language models' (VidLM) impressive performance on various benchmark tasks, our diagnostic tasks reveal their surprising deficiency (near-random performance) in action knowledge, suggesting that current models rely on object recognition abilities as a shortcut for action understanding. To remedy this, we propose a novel framework, Paxion, along with a new Discriminative Video Dynamics Modeling (DVDM) objective. The Paxion framework utilizes a Knowledge Patcher network to encode new action knowledge and a Knowledge Fuser component to integrate the Patcher into frozen VidLMs without compromising their existing capabilities. Due to limitations of the widely-used Video-Text Contrastive (VTC) loss for learning action knowledge, we introduce the DVDM objective to train the Knowledge Patcher. DVDM forces the model to encode the correlation between the action text and the correct ordering of video frames. Our extensive analyses show that Paxion and DVDM together effectively fill the gap in action knowledge understanding (~50% to 80%), while maintaining or improving performance on a wide spectrum of both object- and action-centric downstream tasks.
Discounted Thompson Sampling for Non-Stationary Bandit Problems
Abstract
Non-stationary multi-armed bandit (NS-MAB) problems have recently received significant attention. NS-MAB are typically modelled in two scenarios: abruptly changing, where reward distributions remain constant for a certain period and change at unknown time steps, and smoothly changing, where reward distributions evolve smoothly based on unknown dynamics. In this paper, we propose Discounted Thompson Sampling (DS-TS) with Gaussian priors to address both non-stationary settings. Our algorithm passively adapts to changes by incorporating a discounted factor into Thompson Sampling. DS-TS method has been experimentally validated, but analysis of the regret upper bound is currently lacking. Under mild assumptions, we show that DS-TS with Gaussian priors can achieve nearly optimal regret bound on the order of $\tilde{O}(\sqrt{TB_T})$ for abruptly changing and $\tilde{O}(T^{\beta})$ for smoothly changing, where $T$ is the number of time steps, $B_T$ is the number of breakpoints, $\beta$ is associated with the smoothly changing environment and $\tilde{O}$ hides the parameters independent of $T$ as well as logarithmic terms. Furthermore, empirical comparisons between DS-TS and other non-stationary bandit algorithms demonstrate its competitive performance. Specifically, when prior knowledge of the maximum expected reward is available, DS-TS has the potential to outperform state-of-the-art algorithms.
Counterfactual Debiasing for Generating Factually Consistent Text Summaries
Abstract
Despite substantial progress in abstractive text summarization to generate fluent and informative texts, the factual inconsistency in the generated summaries remains an important yet challenging problem to be solved. In this paper, we construct causal graphs for abstractive text summarization and identify the intrinsic causes of the factual inconsistency, i.e., the language bias and irrelevancy bias, and further propose a debiasing framework, named CoFactSum, to alleviate the causal effects of these biases by counterfactual estimation. Specifically, the proposed CoFactSum provides two counterfactual estimation strategies, i.e., Explicit Counterfactual Masking with an explicit dynamic masking strategy, and Implicit Counterfactual Training with an implicit discriminative cross-attention mechanism. Meanwhile, we design a Debiasing Degree Adjustment mechanism to dynamically adapt the debiasing degree at each decoding step. Extensive experiments on two widely-used summarization datasets demonstrate the effectiveness of CoFactSum in enhancing the factual consistency of generated summaries compared with several baselines.
Deep Temporal Graph Clustering
Authors: Meng Liu, Yue Liu, Ke Liang, Siwei Wang, Sihang Zhou, Xinwang Liu
Abstract
Deep graph clustering has recently received significant attention due to its ability to enhance the representation learning capabilities of models in unsupervised scenarios. Nevertheless, deep clustering for temporal graphs, which could capture crucial dynamic interaction information, has not been fully explored. It means that in many clustering-oriented real-world scenarios, temporal graphs can only be processed as static graphs. This not only causes the loss of dynamic information but also triggers huge computational consumption. To solve the problem, we propose a general framework for deep Temporal Graph Clustering called TGC, which adjusts deep clustering techniques (clustering assignment distribution and adjacency matrix reconstruction) to suit the interaction sequence-based batch-processing pattern of temporal graphs. In addition, we discuss differences between temporal graph clustering and existing static graph clustering from several levels. To verify the superiority of the proposed framework TGC, we conduct extensive experiments. The experimental results show that temporal graph clustering enables more flexibility in finding a balance between time and space requirements, and our framework can effectively improve the performance of existing temporal graph learning methods. Our code and supplementary material will be released after publication.
Latent Space Planning for Multi-Object Manipulation with Environment-Aware Relational Classifiers
Abstract
Objects rarely sit in isolation in everyday human environments. If we want robots to operate and perform tasks in our human environments, they must understand how the objects they manipulate will interact with structural elements of the environment for all but the simplest of tasks. As such, we'd like our robots to reason about how multiple objects and environmental elements relate to one another and how those relations may change as the robot interacts with the world. We examine the problem of predicting inter-object and object-environment relations between previously unseen objects and novel environments purely from partial-view point clouds. Our approach enables robots to plan and execute sequences to complete multi-object manipulation tasks defined from logical relations. This removes the burden of providing explicit, continuous object states as goals to the robot. We explore several different neural network architectures for this task. We find the best performing model to be a novel transformer-based neural network that both predicts object-environment relations and learns a latent-space dynamics function. We achieve reliable sim-to-real transfer without any fine-tuning. Our experiments show that our model understands how changes in observed environmental geometry relate to semantic relations between objects. We show more videos on our website: https://sites.google.com/view/erelationaldynamics.
Best-Response Dynamics in Lottery Contests
Authors: Abheek Ghosh, Paul W. Goldberg
Subjects: Computer Science and Game Theory (cs.GT); Theoretical Economics (econ.TH)
Abstract
We study the convergence of best-response dynamics in lottery contests. We show that best-response dynamics rapidly converges to the (unique) equilibrium for homogeneous agents but may not converge for non-homogeneous agents, even for two non-homogeneous agents. For $2$ homogeneous agents, we show convergence to an $\epsilon$-approximate equilibrium in $\Theta(\log\log(1/\epsilon))$ steps. For $n \ge 3$ agents, the dynamics is not unique because at each step $n-1 \ge 2$ agents can make non-trivial moves. We consider a model where the agent making the move is randomly selected at each time step. We show convergence to an $\epsilon$-approximate equilibrium in $O(\beta \log(n/(\epsilon\delta)))$ steps with probability $1-\delta$, where $\beta$ is a parameter of the agent selection process, e.g., $\beta = n$ if agents are selected uniformly at random at each time step. Our simulations indicate that this bound is tight.
A Generalist Dynamics Model for Control
Authors: Ingmar Schubert, Jingwei Zhang, Jake Bruce, Sarah Bechtle, Emilio Parisotto, Martin Riedmiller, Jost Tobias Springenberg, Arunkumar Byravan, Leonard Hasenclever, Nicolas Heess
Abstract
We investigate the use of transformer sequence models as dynamics models (TDMs) for control. In a number of experiments in the DeepMind control suite, we find that first, TDMs perform well in a single-environment learning setting when compared to baseline models. Second, TDMs exhibit strong generalization capabilities to unseen environments, both in a few-shot setting, where a generalist model is fine-tuned with small amounts of data from the target environment, and in a zero-shot setting, where a generalist model is applied to an unseen environment without any further training. We further demonstrate that generalizing system dynamics can work much better than generalizing optimal behavior directly as a policy. This makes TDMs a promising ingredient for a foundation model of control.
Lyapunov-Driven Deep Reinforcement Learning for Edge Inference Empowered by Reconfigurable Intelligent Surfaces
Authors: Kyriakos Stylianopoulos, Mattia Merluzzi, Paolo Di Lorenzo, George C. Alexandropoulos
Subjects: Information Theory (cs.IT); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
Abstract
In this paper, we propose a novel algorithm for energy-efficient, low-latency, accurate inference at the wireless edge, in the context of 6G networks endowed with reconfigurable intelligent surfaces (RISs). We consider a scenario where new data are continuously generated/collected by a set of devices and are handled through a dynamic queueing system. Building on the marriage between Lyapunov stochastic optimization and deep reinforcement learning (DRL), we devise a dynamic learning algorithm that jointly optimizes the data compression scheme, the allocation of radio resources (i.e., power, transmission precoding), the computation resources (i.e., CPU cycles), and the RIS reflectivity parameters (i.e., phase shifts), with the aim of performing energy-efficient edge classification with end-to-end (E2E) delay and inference accuracy constraints. The proposed strategy enables dynamic control of the system and of the wireless propagation environment, performing a low-complexity optimization on a per-slot basis while dealing with time-varying radio channels and task arrivals, whose statistics are unknown. Numerical results assess the performance of the proposed RIS-empowered edge inference strategy in terms of trade-off between energy, delay, and accuracy of a classification task.
A Bioinspired Synthetic Nervous System Controller for Pick-and-Place Manipulation
Authors: Yanjun Li, Ravesh Sukhnandan, Jeffrey P. Gill, Hillel J. Chiel, Victoria Webster-Wood, Roger D. Quinn
Subjects: Robotics (cs.RO); Systems and Control (eess.SY)
Abstract
The Synthetic Nervous System (SNS) is a biologically inspired neural network (NN). Due to its capability of capturing complex mechanisms underlying neural computation, an SNS model is a candidate for building compact and interpretable NN controllers for robots. Previous work on SNSs has focused on applying the model to the control of legged robots and the design of functional subnetworks (FSNs) to realize dynamical systems. However, the FSN approach has previously relied on the analytical solution of the governing equations, which is difficult for designing more complex NN controllers. Incorporating plasticity into SNSs and using learning algorithms to tune the parameters offers a promising solution for systematic design in this situation. In this paper, we theoretically analyze the computational advantages of SNSs compared with other classical artificial neural networks. We then use learning algorithms to develop compact subnetworks for implementing addition, subtraction, division, and multiplication. We also combine the learning-based methodology with a bioinspired architecture to design an interpretable SNS for the pick-and-place control of a simulated gantry system. Finally, we show that the SNS controller is successfully transferred to a real-world robotic platform without further tuning of the parameters, verifying the effectiveness of our approach.
A Virtual Reality Teleoperation Interface for Industrial Robot Manipulators
Abstract
We address the problem of teleoperating an industrial robot manipulator via a commercially available Virtual Reality (VR) interface. Previous works on VR teleoperation for robot manipulators focus primarily on collaborative or research robot platforms (whose dynamics and constraints differ from industrial robot arms), or only address tasks where the robot's dynamics are not as important (e.g: pick and place tasks). We investigate the usage of commercially available VR interfaces for effectively teleoeprating industrial robot manipulators in a variety of contact-rich manipulation tasks. We find that applying standard practices for VR control of robot arms is challenging for industrial platforms because torque and velocity control is not exposed, and position control is mediated through a black-box controller. To mitigate these problems, we propose a simplified filtering approach to process command signals to enable operators to effectively teleoperate industrial robot arms with VR interfaces in dexterous manipulation tasks. We hope our findings will help robot practitioners implement and setup effective VR teleoperation interfaces for robot manipulators. The proposed method is demonstrated on a variety of contact-rich manipulation tasks which can also involve very precise movement of the robot during execution (videos can be found at https://www.youtube.com/watch?v=OhkCB9mOaBc)
Reinforcement Learning for Legged Robots: Motion Imitation from Model-Based Optimal Control
Authors: AJ Miller, Shamel Fahmi, Matthew Chignoli, Sangbae Kim
Abstract
We propose MIMOC: Motion Imitation from Model-Based Optimal Control. MIMOC is a Reinforcement Learning (RL) controller that learns agile locomotion by imitating reference trajectories from model-based optimal control. MIMOC mitigates challenges faced by other motion imitation RL approaches because the references are dynamically consistent, require no motion retargeting, and include torque references. Hence, MIMOC does not require fine-tuning. MIMOC is also less sensitive to modeling and state estimation inaccuracies than model-based controllers. We validate MIMOC on the Mini-Cheetah in outdoor environments over a wide variety of challenging terrain, and on the MIT Humanoid in simulation. We show cases where MIMOC outperforms model-based optimal controllers, and show that imitating torque references improves the policy's performance.
Sharing Lifelong Reinforcement Learning Knowledge via Modulating Masks
Abstract
Lifelong learning agents aim to learn multiple tasks sequentially over a lifetime. This involves the ability to exploit previous knowledge when learning new tasks and to avoid forgetting. Modulating masks, a specific type of parameter isolation approach, have recently shown promise in both supervised and reinforcement learning. While lifelong learning algorithms have been investigated mainly within a single-agent approach, a question remains on how multiple agents can share lifelong learning knowledge with each other. We show that the parameter isolation mechanism used by modulating masks is particularly suitable for exchanging knowledge among agents in a distributed and decentralized system of lifelong learners. The key idea is that the isolation of specific task knowledge to specific masks allows agents to transfer only specific knowledge on-demand, resulting in robust and effective distributed lifelong learning. We assume fully distributed and asynchronous scenarios with dynamic agent numbers and connectivity. An on-demand communication protocol ensures agents query their peers for specific masks to be transferred and integrated into their policies when facing each task. Experiments indicate that on-demand mask communication is an effective way to implement distributed lifelong reinforcement learning and provides a lifelong learning benefit with respect to distributed RL baselines such as DD-PPO, IMPALA, and PPO+EWC. The system is particularly robust to connection drops and demonstrates rapid learning due to knowledge exchange.
The Dilemma of Choice: Addressing Constraint Selection for Autonomous Robotic Agents
Abstract
The tasks that an autonomous agent is expected to perform are often optional or are incompatible with each other owing to the agent's limited actuation capabilities, specifically the dynamics and control input bounds. We encode tasks as time-dependent state constraints and leverage the advances in multi-objective optimization to formulate the problem of choosing tasks as selection of a feasible subset of constraints that can be satisfied for all time and maximizes a performance metric. We show that this problem, although amenable to reachability or mixed integer model predictive control-based analysis in the offline phase, is NP-Hard in general and therefore requires heuristics to be solved efficiently. When incompatibility in constraints is observed under a given policy that imposes task constraints at each time step in an optimization problem, we assign a Lagrange score to each of these constraints based on the variation in the corresponding Lagrange multipliers over the compatible time horizon. These scores are then used to decide the order in which constraints are dropped in a greedy strategy. We further employ a genetic algorithm to improve upon the greedy strategy. We evaluate our method on a robot waypoint following task when the low-level controllers that impose state constraints are described by Control Barrier Function-based Quadratic Programs and provide a comparison with waypoint selection based on knowledge of backward reachable sets.
Uncertainty Guided Label Denoising for Document-level Distant Relation Extraction
Authors: Qi Sun, Kun Huang, Xiaocui Yang, Pengfei Hong, Kun Zhang, Soujanya Poria
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
Document-level relation extraction (DocRE) aims to infer complex semantic relations among entities in a document. Distant supervision (DS) is able to generate massive auto-labeled data, which can improve DocRE performance. Recent works leverage pseudo labels generated by the pre-denoising model to reduce noise in DS data. However, unreliable pseudo labels bring new noise, e.g., adding false pseudo labels and losing correct DS labels. Therefore, how to select effective pseudo labels to denoise DS data is still a challenge in document-level distant relation extraction. To tackle this issue, we introduce uncertainty estimation technology to determine whether pseudo labels can be trusted. In this work, we propose a Document-level distant Relation Extraction framework with Uncertainty Guided label denoising, UGDRE. Specifically, we propose a novel instance-level uncertainty estimation method, which measures the reliability of the pseudo labels with overlapping relations. By further considering the long-tail problem, we design dynamic uncertainty thresholds for different types of relations to filter high-uncertainty pseudo labels. We conduct experiments on two public datasets. Our framework outperforms strong baselines by 1.91 F1 and 2.28 Ign F1 on the RE-DocRED dataset.
Late-Binding Scholarship in the Age of AI: Navigating Legal and Normative Challenges of a New Form of Knowledge Production
Authors: Bill Tomlinson, Andrew W. Torrance, Rebecca W. Black, Donald J. Patterson
Subjects: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
Abstract
Artificial Intelligence (AI) is poised to enable a new leap in the creation of scholarly content. New forms of engagement with AI systems, such as collaborations with large language models like GPT-3, offer affordances that will change the nature of both the scholarly process and the artifacts it produces. This article articulates ways in which those artifacts can be written, distributed, read, organized, and stored that are more dynamic, and potentially more effective, than current academic practices. Specifically, rather than the current "early-binding" process (that is, one in which ideas are fully reduced to a final written form before they leave an author's desk), we propose that there are substantial benefits to a "late-binding" process, in which ideas are written dynamically at the moment of reading. In fact, the paradigm of "binding" knowledge may transition to a new model in which scholarship remains ever "unbound" and evolving. An alternative form for a scholarly work could be encapsulated via several key components: a text abstract of the work's core arguments; hyperlinks to a bibliography of relevant related work; novel data that had been collected and metadata describing those data; algorithms or processes necessary for analyzing those data; a reference to a particular AI model that would serve as a "renderer" of the canonical version of the text; and specified parameters that would allow for a precise, word-for-word reconstruction of the canonical version. Such a form would enable both the rendering of the canonical version, and also the possibility of dynamic AI reimaginings of the text in light of future findings, scholarship unknown to the original authors, alternative theories, and precise tailoring to specific audiences (e.g., children, adults, professionals, amateurs).
ORKG-Leaderboards: A Systematic Workflow for Mining Leaderboards as a Knowledge Graph
Authors: Salomon Kabongo, Jennifer D'Souza, Sören Auer
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
The purpose of this work is to describe the Orkg-Leaderboard software designed to extract leaderboards defined as Task-Dataset-Metric tuples automatically from large collections of empirical research papers in Artificial Intelligence (AI). The software can support both the main workflows of scholarly publishing, viz. as LaTeX files or as PDF files. Furthermore, the system is integrated with the Open Research Knowledge Graph (ORKG) platform, which fosters the machine-actionable publishing of scholarly findings. Thus the system output, when integrated within the ORKG's supported Semantic Web infrastructure of representing machine-actionable 'resources' on the Web, enables: 1) broadly, the integration of empirical results of researchers across the world, thus enabling transparency in empirical research with the potential to also being complete contingent on the underlying data source(s) of publications; and 2) specifically, enables researchers to track the progress in AI with an overview of the state-of-the-art (SOTA) across the most common AI tasks and their corresponding datasets via dynamic ORKG frontend views leveraging tables and visualization charts over the machine-actionable data. Our best model achieves performances above 90% F1 on the \textit{leaderboard} extraction task, thus proving Orkg-Leaderboards a practically viable tool for real-world usage. Going forward, in a sense, Orkg-Leaderboards transforms the leaderboard extraction task to an automated digitalization task, which has been, for a long time in the community, a crowdsourced endeavor.
Design of the Impulsive Goodwin's Oscillator: A Case Study
Authors: Alexander Medvedev, Anton V. Proskurnikov, Zhanybai T. Zhusubaliyev
Subjects: Systems and Control (eess.SY); Optimization and Control (math.OC)
Abstract
The impulsive Goodwin's oscillator (IGO) is a hybrid model composed of a third-order continuous linear part and a pulse-modulated feedback. This paper introduces a design problem of the IGO to admit a desired periodic solution. The dynamics of the continuous states represent the plant to be controlled, whereas the parameters of the impulsive feedback constitute design degrees of freedom. The design objective is to select the free parameters so that the IGO exhibits a stable 1-cycle with desired characteristics. The impulse-to-impulse map of the oscillator is demonstrated to always possess a positive fixed point that corresponds to the desired periodic solution; the closed-form expressions to evaluate this fixed point are provided. Necessary and sufficient conditions for orbital stability of the 1-cycle are presented in terms of the oscillator parameters and exhibit similarity to the problem of static output control. An IGO design procedure is proposed and validated by simulation. The nonlinear dynamics of the designed IGO are reviewed by means of bifurcation analysis. Applications of the design procedure to dosing problems in chemical industry and biomedicine are envisioned.
Efficient Prompting via Dynamic In-Context Learning
Authors: Wangchunshu Zhou, Yuchen Eleanor Jiang, Ryan Cotterell, Mrinmaya Sachan
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Abstract
The primary way of building AI applications is shifting from training specialist models to prompting generalist models. A common practice for prompting generalist models, often referred to as in-context learning, is to append a few examples (demonstrations) to the prompt to help the model better understand the task. While effective, in-context learning can be inefficient because it makes the input prompt much longer, consuming valuable space in the context window and leading to larger computational costs. In this paper, we propose DynaICL, a recipe for efficient prompting with black-box generalist models that dynamically allocate in-context examples according to the input complexity and the computational budget. To achieve this, we train a meta controller that predicts the number of in-context examples suitable for the generalist model to make a good prediction based on the performance-efficiency trade-off for a specific input. We then dynamically allocate the number of demonstrations for an input according to predictions from the meta controller and the given computation budget. Experimental results show that dynamic example allocation helps achieve a better performance-efficiency trade-off in two practical settings where computational resources or the required performance is constrained. Specifically, DynaICL saves up to 46% token budget compared to the common practice that allocates the same number of in-context examples to each input. We also find that a meta controller trained on a certain backbone model and tasks can successfully generalize to unseen models and tasks.
Keyword: efficient
An Allocation Model for Attributing Emissions in Multi-tenant Cloud Data Centers
An Intelligent SDWN Routing Algorithm Based on Network Situational Awareness and Deep Reinforcement Learning
Optimizing Forest Fire Prevention: Intelligent Scheduling Algorithms for Drone-Based Surveillance System
AnalogNAS: A Neural Network Design Framework for Accurate Inference with Analog In-Memory Computing
Statistical Knowledge Assessment for Generative Language Models
Integrated Conflict Management for UAM with Strategic Demand Capacity Balancing and Learning-based Tactical Deconfliction
The Complexity of Diagonalization
MultiPlaneNeRF: Neural Radiance Field with Non-Trainable Representation
Unsourced Massive Access-Based Digital Over-the-Air Computation for Efficient Federated Edge Learning
ACRoBat: Optimizing Auto-batching of Dynamic Deep Learning at Compile Time
Accelerating MPI Collectives with Process-in-Process-based Multi-object Techniques
TSoR: TCP Socket over RDMA Container Network for Cloud Native Computing
Smoothing the Landscape Boosts the Signal for SGD: Optimal Sample Complexity for Learning Single Index Models
Incremental Causal Graph Learning for Online Unsupervised Root Cause Analysis
PTQD: Accurate Post-Training Quantization for Diffusion Models
Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learning
Zero-Day Backdoor Attack against Text-to-Image Diffusion Models via Personalization
Extracting Low-/High- Frequency Knowledge from Graph Neural Networks and Injecting it into MLPs: An Effective GNN-to-MLP Distillation Framework
OpenShape: Scaling Up 3D Shape Representation Towards Open-World Understanding
Ditto: A Simple and Efficient Approach to Improve Sentence Embeddings
Joint BS Mode Selection and Beamforming Design for Cooperative Cell-Free ISAC Networks
Two-step Newton's method for deflation-one singular zeros of analytic systems
Adaptive choice of near-optimal expansion points for interpolation-based structure-preserving model reduction
GraphMoco:a Graph Momentum Contrast Model that Using Multimodel Structure Information for Large-scale Binary Function Representation Learning
Ahead-of-Time P-Tuning
X-IQE: eXplainable Image Quality Evaluation for Text-to-Image Generation with Visual Large Language Models
Q-SHED: Distributed Optimization at the Edge via Hessian Eigenvectors Quantization
FLIGHT Mode On: A Feather-Light Network for Low-Light Image Enhancement
EventNet-ITA: Italian Frame Parsing for Events
Ultra-High Resolution Segmentation with Ultra-Rich Context: A Novel Benchmark
Structural Pruning for Diffusion Models
Unsupervised Pansharpening via Low-rank Diffusion Model
Lyapunov-Driven Deep Reinforcement Learning for Edge Inference Empowered by Reconfigurable Intelligent Surfaces
Toward Platform-based Building Design
Making More of Little Data: Improving Low-Resource Automatic Speech Recognition Using Data Augmentation
Learning Activation Functions for Sparse Neural Networks
Stopping Criteria for the Conjugate Gradient Algorithm in High-Order Finite Element Methods
Near-Field 3D Localization via MIMO Radar: Cramér-Rao Bound and Estimator Design
How does the task complexity of masked pretraining objectives affect downstream performance?
Mode Connectivity in Auction Design
The Dilemma of Choice: Addressing Constraint Selection for Autonomous Robotic Agents
Generalized Planning in PDDL Domains with Pretrained Large Language Models
Optimistic Natural Policy Gradient: a Simple Efficient Policy Optimization Framework for Online RL
SPSQL: Step-by-step Parsing Based Framework for Text-to-SQL Generation
Blendstrings: an environment for computing with smooth functions
Blockwise inversion and algorithms for inverting large partitioned matrices
Using Symbolic Computation to Analyze Zero-Hopf Bifurcations of Polynomial Differential Systems
mLongT5: A Multilingual and Efficient Text-To-Text Transformer for Longer Sequences
SimOAP: Improve Coherence and Consistency in Persona-based Dialogue Generation via Over-sampling and Post-evaluation
Convergence Analysis of Over-the-Air FL with Compression and Power Control via Clipping
Exploring the Carbon Footprint of Hugging Face's ML Models: A Repository Mining Study
Efficient Prompting via Dynamic In-Context Learning
Keyword: faster
RAMP: Hierarchical Reactive Motion Planning for Manipulation Tasks Using Implicit Signed Distance Functions
Democratized Diffusion Language Model
A Lexical-aware Non-autoregressive Transformer-based ASR Model
TAPIR: Learning Adaptive Revision for Incremental Natural Language Understanding with a Two-Pass Model
Benchmark Framework with Skewed Workloads
Hibernate Container: A Deflated Container Mode for Fast Startup and High-density Deployment in Serverless Computing
Deep PackGen: A Deep Reinforcement Learning Framework for Adversarial Network Packet Generation
MVPSNet: Fast Generalizable Multi-view Photometric Stereo
Keyword: mobile
Sim-MEES: Modular End-Effector System Grasping Dataset for Mobile Manipulators in Cluttered Environments
Multi-microservice migration modelling, comparison, and potential in 5G/6G mobile edge computing: A non-average parameter values approach
Robust Single-Point Pushing with Force Feedback
XFormer: Fast and Accurate Monocular 3D Body Capture
Keyword: pruning
Less Can Be More: Unsupervised Graph Pruning for Large-scale Dynamic Graphs
Boost Vision Transformer with GPU-Friendly Sparsity and Quantization
Structural Pruning for Diffusion Models
Learning Activation Functions for Sparse Neural Networks
Keyword: voxel
There is no result
Keyword: lidar
Improving Extrinsics between RADAR and LIDAR using Learning
Keyword: diffusion
Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models
PTQD: Accurate Post-Training Quantization for Diffusion Models
Content-based Unrestricted Adversarial Attack
RMSSinger: Realistic-Music-Score based Singing Voice Synthesis
Sampling, Diffusions, and Stochastic Localization
Dirichlet Diffusion Score Model for Biological Sequence Generation
Zero-Day Backdoor Attack against Text-to-Image Diffusion Models via Personalization
Discriminative Diffusion Models as Few-shot Vision and Language Learners
Diffusion-Based Speech Enhancement with Joint Generative and Predictive Decoders
Multi-resolution Spatiotemporal Enhanced Transformer Denoising with Functional Diffusive GANs for Constructing Brain Effective Connectivity in MCI analysis
Catch-Up Distillation: You Only Need to Train Once for Accelerating Sampling
Supercloseness of the LDG method for a two-dimensional singularly perturbed convection-diffusion problem on Bakhvalov-type mesh
Democratized Diffusion Language Model
DiffUTE: Universal Text Editing Diffusion Model
Constructing a personalized AI assistant for shear wall layout using Stable Diffusion
AIwriting: Relations Between Image Generation and Digital Writing
GETMusic: Generating Any Music Tracks with a Unified Representation and Diffusion Framework
X-IQE: eXplainable Image Quality Evaluation for Text-to-Image Generation with Visual Large Language Models
LDM3D: Latent Diffusion Model for 3D
TextDiffuser: Diffusion Models as Text Painters
Structural Pruning for Diffusion Models
Unsupervised Pansharpening via Low-rank Diffusion Model
Generating coherent comic with rich story using ChatGPT and Stable Diffusion
Inspecting the Geographical Representativeness of Images from Text-to-Image Models
Blackout Diffusion: Generative Diffusion Models in Discrete-State Spaces
UniControl: A Unified Diffusion Model for Controllable Visual Generation In the Wild
Keyword: dynamic
Bringing AI to the edge: A formal M&S specification to deploy effective IoT architectures
Intelligent multicast routing method based on multi-agent deep reinforcement learning in SDWN
An Intelligent SDWN Routing Algorithm Based on Network Situational Awareness and Deep Reinforcement Learning
Emotion Recognition based on Psychological Components in Guided Narratives for Emotion Regulation
The Effectiveness of a Dynamic Loss Function in Neural Network Based Automated Essay Scoring
Exact Recovery for System Identification with More Corrupt Data than Clean Data
Analytic relationship of relative synchronizability to network structure and motifs
On a Doubly Reduced Model for Dynamics of Heterogeneous Mixtures of Stiffened Gases, its Regularizations and their Implementations
RAMP: Hierarchical Reactive Motion Planning for Manipulation Tasks Using Implicit Signed Distance Functions
Discovering Individual Rewards in Collective Behavior through Inverse Multi-Agent Reinforcement Learning
Sim-MEES: Modular End-Effector System Grasping Dataset for Mobile Manipulators in Cluttered Environments
ACRoBat: Optimizing Auto-batching of Dynamic Deep Learning at Compile Time
Learning Differentially Private Probabilistic Models for Privacy-Preserving Image Generation
Relay Mining: Verifiable Multi-Tenant Distributed Rate Limiting
Less Can Be More: Unsupervised Graph Pruning for Large-scale Dynamic Graphs
Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learning
Paxion: Patching Action Knowledge in Video-Language Foundation Models
Discounted Thompson Sampling for Non-Stationary Bandit Problems
Counterfactual Debiasing for Generating Factually Consistent Text Summaries
Deep Temporal Graph Clustering
Latent Space Planning for Multi-Object Manipulation with Environment-Aware Relational Classifiers
Best-Response Dynamics in Lottery Contests
A Generalist Dynamics Model for Control
Lyapunov-Driven Deep Reinforcement Learning for Edge Inference Empowered by Reconfigurable Intelligent Surfaces
A Bioinspired Synthetic Nervous System Controller for Pick-and-Place Manipulation
A Virtual Reality Teleoperation Interface for Industrial Robot Manipulators
Reinforcement Learning for Legged Robots: Motion Imitation from Model-Based Optimal Control
Sharing Lifelong Reinforcement Learning Knowledge via Modulating Masks
The Dilemma of Choice: Addressing Constraint Selection for Autonomous Robotic Agents
Uncertainty Guided Label Denoising for Document-level Distant Relation Extraction
Late-Binding Scholarship in the Age of AI: Navigating Legal and Normative Challenges of a New Form of Knowledge Production
ORKG-Leaderboards: A Systematic Workflow for Mining Leaderboards as a Knowledge Graph
Design of the Impulsive Goodwin's Oscillator: A Case Study
Efficient Prompting via Dynamic In-Context Learning