{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-efficient-convolutional-networks","title":"Learning Efficient Convolutional Networks through Network Slimming","arxiv_id":"1708.06519","date":"2017-08-22","proceeding":"ICCV 2017 10","authors":["Zhuang Liu","Jianguo Li","Zhiqiang Shen","Gao Huang","Shoumeng Yan","Chang-Shui Zhang"],"abstract":"The deployment of deep convolutional neural networks (CNNs) in many real\nworld applications is largely hindered by their high computational cost. In\nthis paper, we propose a novel learning scheme for CNNs to simultaneously 1)\nreduce the model size; 2) decrease the run-time memory footprint; and 3) lower\nthe number of computing operations, without compromising accuracy. This is\nachieved by enforcing channel-level sparsity in the network in a simple but\neffective way. Different from many existing approaches, the proposed method\ndirectly applies to modern CNN architectures, introduces minimum overhead to\nthe training process, and requires no special software/hardware accelerators\nfor the resulting models. We call our approach network slimming, which takes\nwide and large networks as input models, but during training insignificant\nchannels are automatically identified and pruned afterwards, yielding thin and\ncompact models with comparable accuracy. We empirically demonstrate the\neffectiveness of our approach with several state-of-the-art CNN models,\nincluding VGGNet, ResNet and DenseNet, on various image classification\ndatasets. For VGGNet, a multi-pass version of network slimming gives a 20x\nreduction in model size and a 5x reduction in computing operations.","url_abs":"http://arxiv.org/abs/1708.06519v1","url_pdf":"http://arxiv.org/pdf/1708.06519v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/liuzhuang13/slimming","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/Eric-mingjie/network-slimming","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/DDGRCF/yolov5_Chinese","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/DDGRCF/yolov5_pruning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/EstherBear/implementation-of-network-slimming","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/VainF/Torch-Pruning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/ZJCV/NetworkSlimming","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/chumingqian/Model_Compression_For_YOLOV3-V4","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/midasklr/yolov5prune","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/mvpzhangqiu/yolov5prune","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/polarizationpruning/PolarizationPruning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-efficient-convolutional-networks","repo_url":"https://github.com/alibaba/esim-response-selection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"architecture-search","task_name":"Neural Architecture Search"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-block","method_name":"Dense Block"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1708.06519","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}