{"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/exponential-moving-average-normalization-for","title":"Exponential Moving Average Normalization for Self-supervised and Semi-supervised Learning","arxiv_id":"2101.08482","date":"2021-01-21","proceeding":"CVPR 2021 1","authors":["Zhaowei Cai","Avinash Ravichandran","Subhransu Maji","Charless Fowlkes","Zhuowen Tu","Stefano Soatto"],"abstract":"We present a plug-in replacement for batch normalization (BN) called exponential moving average normalization (EMAN), which improves the performance of existing student-teacher based self- and semi-supervised learning techniques. Unlike the standard BN, where the statistics are computed within each batch, EMAN, used in the teacher, updates its statistics by exponential moving average from the BN statistics of the student. This design reduces the intrinsic cross-sample dependency of BN and enhances the generalization of the teacher. EMAN improves strong baselines for self-supervised learning by 4-6/1-2 points and semi-supervised learning by about 7/2 points, when 1%/10% supervised labels are available on ImageNet. These improvements are consistent across methods, network architectures, training duration, and datasets, demonstrating the general effectiveness of this technique. The code is available at https://github.com/amazon-research/exponential-moving-average-normalization.","url_abs":"https://arxiv.org/abs/2101.08482v2","url_pdf":"https://arxiv.org/pdf/2101.08482v2.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":"exponential-moving-average-normalization-for","repo_url":"https://github.com/amazon-research/exponential-moving-average-normalization","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"self-supervised-learning","task_name":"Self-Supervised Learning"},{"task_slug":"semi-supervised-image-classification","task_name":"Semi-Supervised Image Classification"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semi-supervised-image-classification-on-16","task":"Semi-Supervised Image Classification","dataset":"ImageNet - 0.2% labeled data","model":"FixMatch w/ EMAN (ResNet-50)","rank_in_archive_order":2,"of":3,"metrics":{"ImageNet Top-1 Accuracy":"43.6%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-image-classification-on-1","task":"Semi-Supervised Image Classification","dataset":"ImageNet - 1% labeled data","model":"FixMatch-EMAN","rank_in_archive_order":34,"of":65,"metrics":{"Top 1 Accuracy":"63%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-image-classification-on-2","task":"Semi-Supervised Image Classification","dataset":"ImageNet - 10% labeled data","model":"FixMatch-EMAN","rank_in_archive_order":31,"of":75,"metrics":{"Top 1 Accuracy":"74%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2101.08482","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}