{"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/the-non-iid-data-quagmire-of-decentralized","title":"The Non-IID Data Quagmire of Decentralized Machine Learning","arxiv_id":"1910.00189","date":"2019-10-01","proceeding":"ICML 2020 1","authors":["Kevin Hsieh","Amar Phanishayee","Onur Mutlu","Phillip B. Gibbons"],"abstract":"Many large-scale machine learning (ML) applications need to perform decentralized learning over datasets generated at different devices and locations. Such datasets pose a significant challenge to decentralized learning because their different contexts result in significant data distribution skew across devices/locations. In this paper, we take a step toward better understanding this challenge by presenting a detailed experimental study of decentralized DNN training on a common type of data skew: skewed distribution of data labels across devices/locations. Our study shows that: (i) skewed data labels are a fundamental and pervasive problem for decentralized learning, causing significant accuracy loss across many ML applications, DNN models, training datasets, and decentralized learning algorithms; (ii) the problem is particularly challenging for DNN models with batch normalization; and (iii) the degree of data skew is a key determinant of the difficulty of the problem. Based on these findings, we present SkewScout, a system-level approach that adapts the communication frequency of decentralized learning algorithms to the (skew-induced) accuracy loss between data partitions. We also show that group normalization can recover much of the accuracy loss of batch normalization.","url_abs":"https://arxiv.org/abs/1910.00189v2","url_pdf":"https://arxiv.org/pdf/1910.00189v2.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":"the-non-iid-data-quagmire-of-decentralized","repo_url":"https://github.com/kevinhsieh/non_iid_dml","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"group-normalization","method_name":"Group Normalization"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1910.00189","atlas_url":"https://app.syntology.ai/?focus=1910.00189","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}