{"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/adam-admm-a-unified-systematic-framework-of","title":"StructADMM: A Systematic, High-Efficiency Framework of Structured Weight Pruning for DNNs","arxiv_id":"1807.11091","date":"2018-07-29","proceeding":null,"authors":["Tianyun Zhang","Shaokai Ye","Kaiqi Zhang","Xiaolong Ma","Ning Liu","Linfeng Zhang","Jian Tang","Kaisheng Ma","Xue Lin","Makan Fardad","Yanzhi Wang"],"abstract":"Weight pruning methods of DNNs have been demonstrated to achieve a good model\npruning rate without loss of accuracy, thereby alleviating the significant\ncomputation/storage requirements of large-scale DNNs. Structured weight pruning\nmethods have been proposed to overcome the limitation of irregular network\nstructure and demonstrated actual GPU acceleration. However, in prior work the\npruning rate (degree of sparsity) and GPU acceleration are limited (to less\nthan 50%) when accuracy needs to be maintained. In this work,we overcome these\nlimitations by proposing a unified, systematic framework of structured weight\npruning for DNNs. It is a framework that can be used to induce different types\nof structured sparsity, such as filter-wise, channel-wise, and shape-wise\nsparsity, as well non-structured sparsity. The proposed framework incorporates\nstochastic gradient descent with ADMM, and can be understood as a dynamic\nregularization method in which the regularization target is analytically\nupdated in each iteration. Without loss of accuracy on the AlexNet model, we\nachieve 2.58X and 3.65X average measured speedup on two GPUs, clearly\noutperforming the prior work. The average speedups reach 3.15X and 8.52X when\nallowing a moderate ac-curacy loss of 2%. In this case the model compression\nfor convolutional layers is 15.0X, corresponding to 11.93X measured CPU\nspeedup. Our experiments on ResNet model and on other data sets like UCF101 and\nCIFAR-10 demonstrate the consistently higher performance of our framework.","url_abs":"http://arxiv.org/abs/1807.11091v3","url_pdf":"http://arxiv.org/pdf/1807.11091v3.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":"adam-admm-a-unified-systematic-framework-of","repo_url":"https://github.com/KaiqiZhang/ADAM-ADMM","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":null,"task_name":"CPU"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"model-compression","task_name":"Model Compression"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"admm","method_name":"ADMM"},{"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":"convolution","method_name":"Convolution"},{"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":"grouped-convolution","method_name":"Grouped Convolution"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"local-response-normalization","method_name":"Local Response Normalization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"pruning","method_name":"Pruning"},{"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":{"syntology_url":"https://syntology.ai/paper/1807.11091","atlas_url":"https://app.syntology.ai/?focus=1807.11091","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}