{"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/expanding-the-reach-of-federated-learning-by","title":"Expanding the Reach of Federated Learning by Reducing Client Resource Requirements","arxiv_id":"1812.07210","date":"2018-12-18","proceeding":"ICLR 2019 5","authors":["Sebastian Caldas","Jakub Konečny","H. Brendan McMahan","Ameet Talwalkar"],"abstract":"Communication on heterogeneous edge networks is a fundamental bottleneck in\nFederated Learning (FL), restricting both model capacity and user\nparticipation. To address this issue, we introduce two novel strategies to\nreduce communication costs: (1) the use of lossy compression on the global\nmodel sent server-to-client; and (2) Federated Dropout, which allows users to\nefficiently train locally on smaller subsets of the global model and also\nprovides a reduction in both client-to-server communication and local\ncomputation. We empirically show that these strategies, combined with existing\ncompression approaches for client-to-server communication, collectively provide\nup to a $14\\times$ reduction in server-to-client communication, a $1.7\\times$\nreduction in local computation, and a $28\\times$ reduction in upload\ncommunication, all without degrading the quality of the final model. We thus\ncomprehensively reduce FL's impact on client device resources, allowing higher\ncapacity models to be trained, and a more diverse set of users to be reached.","url_abs":"http://arxiv.org/abs/1812.07210v2","url_pdf":"http://arxiv.org/pdf/1812.07210v2.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":"expanding-the-reach-of-federated-learning-by","repo_url":"https://github.com/k1l1/SLT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"federated-learning","task_name":"Federated Learning"}],"methods":[{"method_slug":"dropout","method_name":"Dropout"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1812.07210","atlas_url":"https://app.syntology.ai/?focus=1812.07210","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}