{"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/restructuring-batch-normalization-to","title":"Restructuring Batch Normalization to Accelerate CNN Training","arxiv_id":"1807.01702","date":"2018-07-04","proceeding":null,"authors":["Wonkyung Jung","Daejin Jung","and Byeongho Kim","Sunjung Lee","Wonjong Rhee","Jung Ho Ahn"],"abstract":"Batch Normalization (BN) has become a core design block of modern\nConvolutional Neural Networks (CNNs). A typical modern CNN has a large number\nof BN layers in its lean and deep architecture. BN requires mean and variance\ncalculations over each mini-batch during training. Therefore, the existing\nmemory access reduction techniques, such as fusing multiple CONV layers, are\nnot effective for accelerating BN due to their inability to optimize mini-batch\nrelated calculations during training. To address this increasingly important\nproblem, we propose to restructure BN layers by first splitting a BN layer into\ntwo sub-layers (fission) and then combining the first sub-layer with its\npreceding CONV layer and the second sub-layer with the following activation and\nCONV layers (fusion). The proposed solution can significantly reduce\nmain-memory accesses while training the latest CNN models, and the experiments\non a chip multiprocessor show that the proposed BN restructuring can improve\nthe performance of DenseNet-121 by 25.7%.","url_abs":"http://arxiv.org/abs/1807.01702v2","url_pdf":"http://arxiv.org/pdf/1807.01702v2.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":"restructuring-batch-normalization-to","repo_url":"https://github.com/wkjung/caffe-bn-restructuring-gpu","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}