High bandwidth requirements are an obstacle for accelerating the training and inference of deep neural networks. Most previous research focuses on reducing the size of kernel maps for inference. We analyze parameter sparsity of six popular convolutional neural networks - AlexNet, MobileNet, ResNet-50, SqueezeNet, TinyNet, and VGG16. Of the networks considered, those using ReLU (AlexNet, SqueezeNet, VGG16) contain a high percentage of 0-valued parameters and can be statically pruned. Networks with Non-ReLU activation functions in some cases may not contain any 0-valued parameters (ResNet-50, TinyNet). We also investigate runtime feature map usage and find that input feature maps comprise the majority of bandwidth requirements when depth-wise convolution and point-wise convolutions used. We introduce dynamic runtime pruning of feature maps and show that 10% of dynamic feature map execution can be removed without loss of accuracy. We then extend dynamic pruning to allow for values within an epsilon of zero and show a further 5% reduction of feature map loading with a 1% loss of accuracy in top-1.
Faculty of Science and Technology, University of Macau, Macau, China（澳门大学科学与技术学院）；School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, China（电子科技大学信息与软件工程学院）
Inference for Deep Neural Networks is increasingly being executed locally on mobile and embedded platforms due to its advantages in latency, privacy and connectivity. Since modern System on Chips typically execute a combination of different and dynamic workloads concurrently, it is challenging to consistently meet inference time/energy budget at runtime because of the local computing resources available to the DNNs vary considerably. To address this challenge, a variety of dynamic DNNs were proposed. However, these works have significant memory overhead, limited runtime recoverable compression…
Pruning Deep Neural Networks (DNNs) is a prominent field of study in the goal of inference runtime acceleration. In this paper, we introduce a novel data-free pruning protocol RED++. Only requiring a trained neural network, and not specific to DNN architecture, we exploit an adaptive data-free scalar hashing which exhibits redundancies among neuron weight values. We study the theoretical and empirical guarantees on the preservation of the accuracy from the hashing as well as the expected pruning ratio resulting from the exploitation of said redundancies. We propose a novel data-free pruning te…
George Mason University（乔治梅森大学）；University of Maryland, Baltimore County（马里兰大学巴尔的摩郡分校）；Microsoft（微软）；Northeastern University（东北大学）
Convolutional Neural Networks (CNNs) achieved great cognitive performance at the expense of considerable computation load. To relieve the computation load, many optimization works are developed to reduce the model redundancy by identifying and removing insignificant model components, such as weight sparsity and filter pruning. However, these works only evaluate model components' static significance with internal parameter information, ignoring their dynamic interaction with external inputs. With per-input feature activation, the model component significance can dynamically change, and thus the…
University of California, San Diego（加州大学圣地亚哥分校）；Google Research（谷歌研究院）
Self-attention is a key enabler of state-of-art accuracy for various transformer-based Natural Language Processing models. This attention mechanism calculates a correlation score for each word with respect to the other words in a sentence. Commonly, only a small subset of words highly correlates with the word under attention, which is only determined at runtime. As such, a significant amount of computation is inconsequential due to low attention scores and can potentially be pruned. The main challenge is finding the threshold for the scores below which subsequent computation will be inconseque…
Robert Bosch GmbH（罗伯特·博世有限公司）；University of Lübeck（吕贝克大学）
Unstructured pruning is well suited to reduce the memory footprint of convolutional neural networks (CNNs), both at training and inference time. CNNs contain parameters arranged in $K \times K$ filters. Standard unstructured pruning (SP) reduces the memory footprint of CNNs by setting filter elements to zero, thereby specifying a fixed subspace that constrains the filter. Especially if pruning is applied before or during training, this induces a strong bias. To overcome this, we introduce interspace pruning (IP), a general tool to improve existing pruning methods. It uses filters represented i…
Convolutional Neural Networks (CNNs) compression is crucial to deploying these models in edge devices with limited resources. Existing channel pruning algorithms for CNNs have achieved plenty of success on complex models. They approach the pruning problem from various perspectives and use different metrics to guide the pruning process. However, these metrics mainly focus on the model's `outputs' or `weights' and neglect its `interpretations' information. To fill in this gap, we propose to address the channel pruning problem from a novel perspective by leveraging the interpretations of a model …
Unstructured neural network pruning algorithms have achieved impressive compression rates. However, the resulting - typically irregular - sparse matrices hamper efficient hardware implementations, leading to additional memory usage and complex control logic that diminishes the benefits of unstructured pruning. This has spurred structured coarse-grained pruning solutions that prune entire filters or even layers, enabling efficient implementation at the expense of reduced flexibility. Here we propose a flexible new pruning mechanism that facilitates pruning at different granularities (weights, k…
Predicting human fixations from images has recently seen large improvements by leveraging deep representations which were pretrained for object recognition. However, as we show in this paper, these networks are highly overparameterized for the task of fixation prediction. We first present a simple yet principled greedy pruning method which we call Fisher pruning. Through a combination of knowledge distillation and Fisher pruning, we obtain much more runtime-efficient architectures for saliency prediction, achieving a 10x speedup for the same AUC performance as a state of the art network on the…
Network pruning has become the de facto tool to accelerate deep neural networks for mobile and edge applications. Recently, feature-map discriminant based channel pruning has shown promising results, as it aligns well with the CNN objective of differentiating multiple classes and offers better interpretability of the pruning decision. However, existing discriminant-based methods are challenged by computation inefficiency, as there is a lack of theoretical guidance on quantifying the feature-map discriminant power. In this paper, we present a new mathematical formulation to accurately and effic…
Northeastern University（东北大学）；College of William and Mary（威廉与玛丽学院）；Michigan State University（密歇根州立大学）；University of Pittsburgh（匹兹堡大学）
Weight pruning is an effective model compression technique to tackle the challenges of achieving real-time deep neural network (DNN) inference on mobile devices. However, prior pruning schemes have limited application scenarios due to accuracy degradation, difficulty in leveraging hardware acceleration, and/or restriction on certain types of DNN layers. In this paper, we propose a general, fine-grained structured pruning scheme and corresponding compiler optimizations that are applicable to any type of DNN layer while achieving high accuracy and hardware inference performance. With the flexibi…
University of Science and Technology Beijing（北京科技大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Convolutional Neural Networks (CNNs) has been applied in numerous Internet of Things (IoT) devices for multifarious downstream tasks. However, with the increasing amount of data on edge devices, CNNs can hardly complete some tasks in time with limited computing and storage resources. Recently, filter pruning has been regarded as an effective technique to compress and accelerate CNNs, but existing methods rarely prune CNNs from the perspective of compressing high-dimensional tensors. In this paper, we propose a novel theory to find redundant information in three-dimensional tensors, namely Quan…
Peking University（北京大学）；Noah’s Ark Lab, Huawei Technologies（华为技术有限公司诺亚方舟实验室）；University of Sydney（悉尼大学）
This paper proposes a reliable neural network pruning algorithm by setting up a scientific control. Existing pruning methods have developed various hypotheses to approximate the importance of filters to the network and then execute filter pruning accordingly. To increase the reliability of the results, we prefer to have a more rigorous research design by including a scientific control group as an essential part to minimize the effect of all factors except the association between the filter and expected network output. Acting as a control group, knockoff feature is generated to mimic the featur…
Bar-Ilan University（巴伊兰大学）；College of Law and Business（法律与商业学院）
The rise of neural network (NN) applications has prompted an increased interest in compression, with a particular focus on channel pruning, which does not require any additional hardware. Most pruning methods employ either single-layer operations or global schemes to determine which channels to remove followed by fine-tuning of the network. In this paper we present Gator, a channel-pruning method which temporarily adds learned gating mechanisms for pruning of individual channels, and which is trained with an additional auxiliary loss, aimed at reducing the computational cost due to memory, (th…
Neural Architecture Search (NAS) has enabled automatic discovery of more efficient neural network architectures, especially for mobile and embedded vision applications. Although recent research has proposed ways of quickly estimating latency on unseen hardware devices with just a few samples, little focus has been given to the challenges of estimating latency on runtimes using optimized graphs, such as TensorRT and specifically for edge devices. In this work, we propose MAPLE-Edge, an edge device-oriented extension of MAPLE, the state-of-the-art latency predictor for general purpose hardware, …
Indian Institute of Information Technology, Sri City（斯里城印度信息技术学院）；Indian Institute of Information Technology, Allahabad（阿拉哈巴德印度信息技术学院）；DryvAmigo（DryvAmigo公司）
This paper proposes an Information Bottleneck theory based filter pruning method that uses a statistical measure called Mutual Information (MI). The MI between filters and class labels, also called \textit{Relevance}, is computed using the filter's activation maps and the annotations. The filters having High Relevance (HRel) are considered to be more important. Consequently, the least important filters, which have lower Mutual Information with the class labels, are pruned. Unlike the existing MI based pruning methods, the proposed method determines the significance of the filters purely based …
Structural pruning can simplify network architecture and improve inference speed. We propose Hardware-Aware Latency Pruning (HALP) that formulates structural pruning as a global resource allocation optimization problem, aiming at maximizing the accuracy while constraining latency under a predefined budget. For filter importance ranking, HALP leverages latency lookup table to track latency reduction potential and global saliency score to gauge accuracy drop. Both metrics can be evaluated very efficiently during pruning, allowing us to reformulate global structural pruning under a reward maximiz…
Xiamen University（厦门大学）；Peng Cheng Laboratory（鹏城实验室）；Pinterest（品趣志）；Beihang University（北京航空航天大学）；Peking University（北京大学）；Inception Institute of Artificial Intelligence（人工智能创想研究院）
Neural network pruning offers a promising prospect to facilitate deploying deep neural networks on resource-limited devices. However, existing methods are still challenged by the training inefficiency and labor cost in pruning designs, due to missing theoretical guidance of non-salient network components. In this paper, we propose a novel filter pruning method by exploring the High Rank of feature maps (HRank). Our HRank is inspired by the discovery that the average rank of multiple feature maps generated by a single filter is always the same, regardless of the number of image batches CNNs rec…
OrderDP: A Theoretically Guaranteed Lossless Dynamic Data Pruning Framework
The Chinese University of Hong Kong（香港中文大学）；Beijing Normal-Hong Kong Baptist University（北京师范大学-香港 Baptist大学）；Guangzhou Nanfang College（广州南方学院）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Xiangtan University（湘潭大学）；University of Utah（犹他大学）
University of the Philippines（菲律宾大学）；Samsung Research Philippines（三星研究院菲律宾分院）
Pruning is a neural network optimization technique that sacrifices accuracy in exchange for lower computational requirements. Pruning has been useful when working with extremely constrained environments in tinyML. Unfortunately, special hardware requirements and limited study on its effectiveness on already compact models prevent its wider adoption. Depth pruning is a form of pruning that requires no specialized hardware but suffers from a large accuracy falloff. To improve this, we propose a modification that utilizes a highly efficient auxiliary network as an effective interpreter of interme…
In recent years, most of the deep learning solutions are targeted to be deployed in mobile devices. This makes the need for development of lightweight models all the more imminent. Another solution is to optimize and prune regular deep learning models. In this paper, we tackle the problem of CNN model pruning with the help of Self-Similarity Matrix (SSM) computed from the 2D CNN filters. We propose two novel algorithms to rank and prune redundant filters which contribute similar activation maps to the output. One of the key features of our method is that there is no need of finetuning after tr…
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Channel pruning, which seeks to reduce the model size by removing redundant channels, is a popular solution for deep networks compression. Existing channel pruning methods usually conduct layer-wise channel selection by directly minimizing the reconstruction error of feature maps between the baseline model and the pruned one. However, they ignore the feature and semantic distributions within feature maps and real contribution of channels to the overall performance. In this paper, we propose a new channel pruning method by explicitly using both intermediate outputs of the baseline model and the…
Convolutional neural network (CNN) has been widely used for vision-based tasks. Due to the high computational complexity and memory storage requirement, it is hard to directly deploy a full-precision CNN on embedded devices. The hardware-friendly designs are needed for re-source-limited and energy-constrained embed-ded devices. Emerging solutions are adopted for the neural network compression, e.g., bina-ry/ternary weight network, pruned network and quantized network. Among them, Binarized Neural Network (BNN) is believed to be the most hardware-friendly framework due to its small network size…
College of Computer Science, Sichuan Normal University（四川师范大学计算机科学学院）；Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）
Channel pruning has demonstrated its effectiveness in compressing ConvNets. In many related arts, the importance of an output feature map is only determined by its associated filter. However, these methods ignore a small part of weights in the next layer which disappears as the feature map is removed. They ignore the phenomenon of weight dependency. Besides, many pruning methods use only one criterion for evaluation and find a sweet spot of pruning structure and accuracy in a trial-and-error fashion, which can be time-consuming. In this paper, we proposed a channel pruning algorithm via multi-…
University of Maryland（马里兰大学）；NEC Labs America（NEC美国实验室）
The success of CNNs in various applications is accompanied by a significant increase in the computation and parameter storage costs. Recent efforts toward reducing these overheads involve pruning and compressing the weights of various layers without hurting original accuracy. However, magnitude-based pruning of weights reduces a significant number of parameters from the fully connected layers and may not adequately reduce the computation costs in the convolutional layers due to irregular sparsity in the pruned networks. We present an acceleration method for CNNs, where we prune filters from CN…
University of California, Santa Barbara（加州大学圣巴巴拉分校）；Tsinghua University（清华大学）
Crossbar architecture based devices have been widely adopted in neural network accelerators by taking advantage of the high efficiency on vector-matrix multiplication (VMM) operations. However, in the case of convolutional neural networks (CNNs), the efficiency is compromised dramatically due to the large amounts of data reuse. Although some mapping methods have been designed to achieve a balance between the execution throughput and resource overhead, the resource consumption cost is still huge while maintaining the throughput. Network pruning is a promising and widely studied leverage to shri…
Filter pruning has been widely used for neural network compression because of its enabled practical acceleration. To date, most of the existing filter pruning works explore the importance of filters via using intra-channel information. In this paper, starting from an inter-channel perspective, we propose to perform efficient filter pruning using Channel Independence, a metric that measures the correlations among different feature maps. The less independent feature map is interpreted as containing less useful information$/$knowledge, and hence its corresponding filter can be pruned without affe…
University of Technology Sydney（悉尼科技大学）；Baidu Research（百度研究院）
Network pruning reduces the computation costs of an over-parameterized network without performance damage. Prevailing pruning algorithms pre-define the width and depth of the pruned networks, and then transfer parameters from the unpruned network to pruned networks. To break the structure limitation of the pruned networks, we propose to apply neural architecture search to search directly for a network with flexible channel and layer sizes. The number of the channels/layers is learned by minimizing the loss of the pruned networks. The feature map of the pruned network is an aggregation of K fea…
Pruning is a promising approach to compress complex deep learning models in order to deploy them on resource-constrained edge devices. However, many existing pruning solutions are based on unstructured pruning, which yields models that cannot efficiently run on commodity hardware and require users to manually explore and tune the pruning process, which is time-consuming and often leads to sub-optimal results. To address these limitations, this paper presents an adaptive, activation-based, structured pruning approach to automatically and efficiently generate small, accurate, and hardware-effici…
Google Research（谷歌研究院）；University of California, San Diego（加州大学圣地亚哥分校）
As its core computation, a self-attention mechanism gauges pairwise correlations across the entire input sequence. Despite favorable performance, calculating pairwise correlations is prohibitively costly. While recent work has shown the benefits of runtime pruning of elements with low attention scores, the quadratic complexity of self-attention mechanisms and their on-chip memory capacity demands are overlooked. This work addresses these constraints by architecting an accelerator, called SPRINT, which leverages the inherent parallelism of ReRAM crossbar arrays to compute attention scores in an…
We propose a simple but effective data-driven channel pruning algorithm, which compresses deep neural networks in a differentiable way by exploiting the characteristics of operations. The proposed approach makes a joint consideration of batch normalization (BN) and rectified linear unit (ReLU) for channel pruning; it estimates how likely the two successive operations deactivate each feature map and prunes the channels with high probabilities. To this end, we learn differentiable masks for individual channels and make soft decisions throughout the optimization procedure, which facilitates to ex…
Hong Kong University of Science and Technology（香港科技大学）；MEGVII Technology（旷视科技）；Carnegie Mellon University（卡内基梅隆大学）
We present joint multi-dimension pruning (abbreviated as JointPruning), an effective method of pruning a network on three crucial aspects: spatial, depth and channel simultaneously. To tackle these three naturally different dimensions, we proposed a general framework by defining pruning as seeking the best pruning vector (i.e., the numerical value of layer-wise channel number, spacial size, depth) and construct a unique mapping from the pruning vector to the pruned network structures. Then we optimize the pruning vector with gradient update and model joint pruning as a numerical gradient optim…
We propose three novel pruning techniques to improve the cost and results of inference-aware Differentiable Neural Architecture Search (DNAS). First, we introduce Prunode, a stochastic bi-path building block for DNAS, which can search over inner hidden dimensions with O(1) memory and compute complexity. Second, we present an algorithm for pruning blocks within a stochastic layer of the SuperNet during the search. Third, we describe a novel technique for pruning unnecessary stochastic layers during the search. The optimized models resulting from the search are called PruNet and establishes a ne…
Structural pruning can simplify network architecture and improve inference speed. We propose Hardware-Aware Latency Pruning (HALP) that formulates structural pruning as a global resource allocation optimization problem, aiming at maximizing the accuracy while constraining latency under a predefined budget on targeting device. For filter importance ranking, HALP leverages latency lookup table to track latency reduction potential and global saliency score to gauge accuracy drop. Both metrics can be evaluated very efficiently during pruning, allowing us to reformulate global structural pruning un…
Can Visual Input Be Compressed? A Visual Token Compression Benchmark for Large Multimodal Models
Shandong University（山东大学）；Shenzhen University of Information Technology（深圳信息职业技术大学）；Tongji University（同济大学）；University of Science and Technology of China（中国科学技术大学）；Sichuan University（四川大学）；University of Notre Dame（圣母大学）；Ludwig Maximilian University of Munich（慕尼黑路德维希-马克西米利安大学）；EB Tech Co., Ltd.（EB科技有限公司）
Isfahan University of Technology（伊斯法罕理工大学）；McMaster University（麦克马斯特大学）
Filters are the essential elements in convolutional neural networks (CNNs). Filters are corresponded to the feature maps and form the main part of the computational and memory requirement for the CNN processing. In filter pruning methods, a filter with all of its components, including channels and connections, are removed. The removal of a filter can cause a drastic change in the network's performance. Also, the removed filters cannot come back to the network structure. We want to address these problems in this paper. We propose a CNN pruning method based on filter attenuation in which weak fi…
We present a filter pruning approach for deep model compression, using a multitask network. Our approach is based on learning a a pruner network to prune a pre-trained target network. The pruner is essentially a multitask deep neural network with binary outputs that help identify the filters from each layer of the original network that do not have any significant contribution to the model and can therefore be pruned. The pruner network has the same architecture as the original network except that it has a multitask/multi-output last layer containing binary-valued outputs (one per filter), whic…
Technical University of Munich, Germany（慕尼黑工业大学，德国）；University of Oxford, UK（牛津大学，英国）；Johns Hopkins University, USA（约翰斯·霍普金斯大学，美国）
Attributing the output of a neural network to the contribution of given input elements is a way of shedding light on the black-box nature of neural networks. Due to the complexity of current network architectures, current gradient-based attribution methods provide very noisy or coarse results. We propose to prune a neural network for a given single input to keep only neurons that highly contribute to the prediction. We show that by input-specific pruning, network gradients change from reflecting local (noisy) importance information to global importance. Our proposed method is efficient and gen…
Structured Pruning for Efficient Visual Place Recognition
School of Electronics and Computer Science, University of Southampton（南安普顿大学电子与计算机科学学院）；School of Electrical Engineering and Computer Science, Queensland University of Technology（昆士兰科技大学电气工程与计算机科学学院）；School of Computer Science and Electronic Engineering, University of Essex（埃塞克斯大学计算机科学与电子工程学院）
Reinforcement Learning-based Token Pruning in Vision Transformers: A Markov Game Approach
School of Computer Science, Fudan University（复旦大学计算机科学学院）；Data Intelligence Institute of Paris（巴黎数据智能研究院）；Université Paris Cité（巴黎西岱大学）；Shijiazhuang Tiedao University（石家庄铁道大学）
This paper presents an efficient technique to prune deep and/or wide convolutional neural network models by eliminating redundant features (or filters). Previous studies have shown that over-sized deep neural network models tend to produce a lot of redundant features that are either shifted version of one another or are very similar and show little or no variations; thus resulting in filtering redundancy. We propose to prune these redundant features along with their connecting feature maps according to their differentiation and based on their relative cosine distances in the feature space, thu…
