{"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/structured-deep-neural-network-pruning-by","title":"Structured Pruning for Efficient ConvNets via Incremental Regularization","arxiv_id":"1804.09461","date":"2018-04-25","proceeding":null,"authors":["Huan Wang","Qiming Zhang","Yuehai Wang","Yu Lu","Haoji Hu"],"abstract":"Parameter pruning is a promising approach for CNN compression and\nacceleration by eliminating redundant model parameters with tolerable\nperformance degrade. Despite its effectiveness, existing regularization-based\nparameter pruning methods usually drive weights towards zero with large and\nconstant regularization factors, which neglects the fragility of the\nexpressiveness of CNNs, and thus calls for a more gentle regularization scheme\nso that the networks can adapt during pruning. To achieve this, we propose a\nnew and novel regularization-based pruning method, named IncReg, to\nincrementally assign different regularization factors to different weights\nbased on their relative importance. Empirical analysis on CIFAR-10 dataset\nverifies the merits of IncReg. Further extensive experiments with popular CNNs\non CIFAR-10 and ImageNet datasets show that IncReg achieves comparable to even\nbetter results compared with state-of-the-arts. Our source codes and trained\nmodels are available here: https://github.com/mingsun-tse/caffe_increg.","url_abs":"http://arxiv.org/abs/1804.09461v2","url_pdf":"http://arxiv.org/pdf/1804.09461v2.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":"structured-deep-neural-network-pruning-by","repo_url":"https://github.com/MingSun-Tse/Caffe_IncReg","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"network-pruning","task_name":"Network Pruning"}],"methods":[{"method_slug":"pruning","method_name":"Pruning"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.09461","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}