Network pruning reduces the computational requirements of large neural networks, with N:M sparsity -- retaining only N out of every M consecutive weights -- offering a compelling balance between compressed model quality and hardware acceleration. However, N:M sparsity only accelerates forward-pass computations, as N:M patterns are not preserved during matrix transposition, limiting efficiency during training where both passes are computationally intensive. While transposable N:M sparsity has been proposed to address this limitation, existing methods for finding transposable N:M sparse masks either fail to scale to large models or are restricted to M=4 which results in suboptimal compression-accuracy trade-off. We introduce an efficient solver for transposable N:M masks that scales to billion-parameter models. We formulate mask generation as optimal transport problems and solve through entropy regularization and Dykstra's algorithm, followed by a rounding procedure. Our tensor-based implementation exploits GPU parallelism, achieving up to 100x speedup with only 1-10% error compared to existing methods. Our approach can be integrated with layer-wise N:M pruning frameworks including Wanda, SparseGPT and ALPS to produce transposable N:M sparse models with arbitrary N:M values. Experiments show that LLaMA3.2-8B with transposable 16:32 sparsity maintains performance close to its standard N:M counterpart and outperforms standard 2:4 sparse model, showing the practical value of our approach.
Kyoto University（京都大学）；National Institute of Informatics（国立信息学研究所）
A common strategy for sparse linear regression is to introduce regularization, which eliminates irrelevant features by letting the corresponding weights be zeros. However, regularization often shrinks the estimator for relevant features, which leads to incorrect feature selection. Motivated by the above-mentioned issue, we propose Bayesian masking (BM), a sparse estimation method which imposes no regularization on the weights. The key concept of BM is to introduce binary latent variables that randomly mask features. Estimating the masking rates determines the relevance of the features automati…
University of Rochester（罗切斯特大学）；Kwai AI Lab（快手人工智能实验室）
Deep Neural Networks (DNNs) are increasingly deployed in highly energy-constrained environments such as autonomous drones and wearable devices while at the same time must operate in real-time. Therefore, reducing the energy consumption has become a major design consideration in DNN training. This paper proposes the first end-to-end DNN training framework that provides quantitative energy consumption guarantees via weighted sparse projection and input masking. The key idea is to formulate the DNN training as an optimization problem in which the energy budget imposes a previously unconsidered op…
With rise of machine learning (ML) and the proliferation of smart mobile devices, recent years have witnessed a surge of interest in performing ML in wireless edge networks. In this paper, we consider the problem of jointly improving data privacy and communication efficiency of distributed edge learning, both of which are critical performance metrics in wireless edge network computing. Toward this end, we propose a new decentralized stochastic gradient method with sparse differential Gaussian-masked stochastic gradients (SDM-DSGD) for non-convex distributed edge learning. Our main contribution…
Xiamen University（厦门大学）；JD Explore Academy（京东探索研究院）；The University of Sydney（悉尼大学）
Deep neural networks often suffer from poor generalization caused by complex and non-convex loss landscapes. One of the popular solutions is Sharpness-Aware Minimization (SAM), which smooths the loss landscape via minimizing the maximized change of training loss when adding a perturbation to the weight. However, we find the indiscriminate perturbation of SAM on all parameters is suboptimal, which also results in excessive computation, i.e., double the overhead of common optimizers like Stochastic Gradient Descent (SGD). In this paper, we propose an efficient and effective training scheme coine…
Peking University（北京大学）；Bytedance Inc.（字节跳动公司）；University of Oxford（牛津大学）
We identify and overcome two key obstacles in extending the success of BERT-style pre-training, or the masked image modeling, to convolutional networks (convnets): (i) convolution operation cannot handle irregular, random-masked input images; (ii) the single-scale nature of BERT pre-training is inconsistent with convnet's hierarchical structure. For (i), we treat unmasked pixels as sparse voxels of 3D point clouds and use sparse convolution to encode. This is the first use of sparse convolution for 2D masked modeling. For (ii), we develop a hierarchical decoder to reconstruct images from multi…
Xiamen University（厦门大学）；Southern Illinois University Carbondale（南伊利诺伊大学卡本代尔分校）；Kneron Inc.（耐能智慧股份有限公司）；Yulin Normal University（玉林师范学院）
Since sparse neural networks usually contain many zero weights, these unnecessary network connections can potentially be eliminated without degrading network performance. Therefore, well-designed sparse neural networks have the potential to significantly reduce FLOPs and computational resources. In this work, we propose a new automatic pruning method - Sparse Connectivity Learning (SCL). Specifically, a weight is re-parameterized as an element-wise multiplication of a trainable weight variable and a binary mask. Thus, network connectivity is fully described by the binary mask, which is modulat…
During typical gradient-based training of deep neural networks, all of the model's parameters are updated at each iteration. Recent work has shown that it is possible to update only a small subset of the model's parameters during training, which can alleviate storage and communication requirements. In this paper, we show that it is possible to induce a fixed sparse mask on the model's parameters that selects a subset to update over many iterations. Our method constructs the mask out of the $k$ parameters with the largest Fisher information as a simple approximation as to which parameters are m…
Foreground detection in a given video sequence is a pivotal step in many computer vision applications such as video surveillance system. Robust Principal Component Analysis (RPCA) performs low-rank and sparse decomposition and accomplishes such a task when the background is stationary and the foreground is dynamic and relatively small. A fundamental issue with RPCA is the assumption that the low-rank and sparse components are added at each element, whereas in reality, the moving foreground is overlaid on the background. We propose the representation via masked decomposition (i.e. an overlaying…
PMA-Diffusion: A Physics-guided Mask-Aware Diffusion Framework for TSE from Sparse Observations
Department of Civil, Environmental, and Geo- Engineering, University of Minnesota, 500 Pillsbury Dr. SE, Minneapolis, MN 55455, USA（土木、环境与地球工程系，明尼苏达大学）；Univ. Gustave Eiffel, ENTPE, EMob-Lab, Lyon, F-69675, France（格勒诺布尔伊夫·欧贝尔大学，ENTPE，EMob-Lab，法国里昂）
SparseSwaps: Tractable LLM Pruning Mask Refinement at Scale
Department for AI in Society, Science, and Technology, Zuse Institute Berlin, Germany（人工智能与社会、科学与技术部门，柏林Zuse研究所，德国）；Institute of Mathematics, Technische Universität Berlin, Germany（数学研究所，柏林技术大学，德国）
The University of Hong Kong（香港大学）；City University of Hong Kong（香港城市大学）
We present a novel network pruning algorithm called Dynamic Sparse Training that can jointly find the optimal network parameters and sparse network structure in a unified optimization process with trainable pruning thresholds. These thresholds can have fine-grained layer-wise adjustments dynamically via backpropagation. We demonstrate that our dynamic sparse training algorithm can easily train very sparse neural network models with little performance loss using the same number of training epochs as dense models. Dynamic Sparse Training achieves the state of the art performance compared with ot…
Diffusion-based inpainting is a powerful tool for the reconstruction of images from sparse data. Its quality strongly depends on the choice of known data. Optimising their spatial location -- the inpainting mask -- is challenging. A commonly used tool for this task are stochastic optimisation strategies. However, they are slow as they compute multiple inpainting results. We provide a remedy in terms of a learned mask generation model. By emulating the complete inpainting pipeline with two networks for mask generation and neural surrogate inpainting, we obtain a model for highly efficient adapt…
Texas A&M University（得克萨斯农工大学）；North Carolina State University（北卡罗来纳州立大学）；Purdue University（普渡大学）
Although sparse training has been successfully used in various resource-limited deep learning tasks to save memory, accelerate training, and reduce inference time, the reliability of the produced sparse models remains unexplored. Previous research has shown that deep neural networks tend to be over-confident, and we find that sparse training exacerbates this problem. Therefore, calibrating the sparse models is crucial for reliable prediction and decision-making. In this paper, we propose a new sparse training method to produce sparse models with improved confidence calibration. In contrast to …
Efficiently Dispatching Flash Attention For Partially Filled Attention Masks
University of Tübingen（蒂宾根大学）；Tübingen AI Center（蒂宾根人工智能中心）；ELLIS Institute Tübingen（ELLIS蒂宾根研究所）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）
Communication-Efficient Federated Learning via Regularized Sparse Random Networks
Mitsubishi Electric R&D Centre Europe（三菱电机欧洲研发中心）；EURECOM（欧洲通信研究中心）；CICT Mobile Communications Technology Co. Ltd.（中国通信建设集团移动通信技术有限公司）；China Academy of Telecommunications Technology (CATT)（中国电信技术研究院）
Deep neural networks (DNNs) have been proven to be vulnerable to adversarial examples. A special branch of adversarial examples, namely sparse adversarial examples, can fool the target DNNs by perturbing only a few pixels. However, many existing sparse adversarial attacks use heuristic methods to select the pixels to be perturbed, and regard the pixel selection and the adversarial attack as two separate steps. From the perspective of neural network pruning, we propose a novel end-to-end sparse adversarial attack method, namely AutoAdversary, which can find the most important pixels automatical…
Task-Specific Sparse Feature Masks for Molecular Toxicity Prediction with Chemical Language Models
School of Science and Technology（科学与技术学院）；Hong Kong Metropolitan University（香港 Metropolitan 大学）；Faculty of Engineering（工程学院）；Hong Kong Polytechnic University（香港理工大学）
Tandon School of Engineering（坦顿工程学院）；New York University（纽约大学）
Deep neural networks are often highly overparameterized, prohibiting their use in compute-limited systems. However, a line of recent works has shown that the size of deep networks can be considerably reduced by identifying a subset of neuron indicators (or mask) that correspond to significant weights prior to training. We demonstrate that an simple iterative mask discovery method can achieve state-of-the-art compression of very deep networks. Our algorithm represents a hybrid approach between single shot network pruning methods (such as SNIP) with Lottery-Ticket type approaches. We validate ou…
Computer-generated holography (CGH) has broad applications such as direct-view display, virtual and augmented reality, as well as optical microscopy. CGH usually utilizes a spatial light modulator that displays a computer-generated phase mask, modulating the phase of coherent light in order to generate customized patterns. The algorithm that computes the phase mask is the core of CGH and is usually tailored to meet different applications. CGH for optical microscopy usually requires 3D accessibility (i.e., generating overlapping patterns along the $z$-axis) and micron-scale spatial precision. H…
SAU: Sparsity-Aware Unlearning for LLMs via Gradient Masking and Importance Redistribution
School of Computer Science and Artificial Intelligence, Wuhan University of Technology（武汉理工大学计算机科学与人工智能学院）；University of California San Diego（加州大学圣地亚哥分校）
Northeastern University（东北大学）；College of William and Mary（威廉与玛丽学院）
Existing work in continual learning (CL) focuses on mitigating catastrophic forgetting, i.e., model performance deterioration on past tasks when learning a new task. However, the training efficiency of a CL system is under-investigated, which limits the real-world application of CL systems under resource-limited scenarios. In this work, we propose a novel framework called Sparse Continual Learning(SparCL), which is the first study that leverages sparsity to enable cost-effective continual learning on edge devices. SparCL achieves both training acceleration and accuracy preservation through the…
Sparsity has become one of the promising methods to compress and accelerate Deep Neural Networks (DNNs). Among different categories of sparsity, structured sparsity has gained more attention due to its efficient execution on modern accelerators. Particularly, N:M sparsity is attractive because there are already hardware accelerator architectures that can leverage certain forms of N:M structured sparsity to yield higher compute-efficiency. In this work, we focus on N:M sparsity and extensively study and evaluate various training recipes for N:M sparsity in terms of the trade-off between model a…
In this paper, we investigate a deep learning approach for speech denoising through an efficient ensemble of specialist neural networks. By splitting up the speech denoising task into non-overlapping subproblems and introducing a classifier, we are able to improve denoising performance while also reducing computational complexity. More specifically, the proposed model incorporates a gating network which assigns noisy speech signals to an appropriate specialist network based on either speech degradation level or speaker gender. In our experiments, a baseline recurrent network is compared agains…
