{"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/subspace-learning-for-personalized-federated","title":"Connecting Low-Loss Subspace for Personalized Federated Learning","arxiv_id":"2109.07628","date":"2021-09-16","proceeding":null,"authors":["Seok-Ju Hahn","Minwoo Jeong","Junghye Lee"],"abstract":"Due to the curse of statistical heterogeneity across clients, adopting a personalized federated learning method has become an essential choice for the successful deployment of federated learning-based services. Among diverse branches of personalization techniques, a model mixture-based personalization method is preferred as each client has their own personalized model as a result of federated learning. It usually requires a local model and a federated model, but this approach is either limited to partial parameter exchange or requires additional local updates, each of which is helpless to novel clients and burdensome to the client's computational capacity. As the existence of a connected subspace containing diverse low-loss solutions between two or more independent deep networks has been discovered, we combined this interesting property with the model mixture-based personalized federated learning method for improved performance of personalization. We proposed SuPerFed, a personalized federated learning method that induces an explicit connection between the optima of the local and the federated model in weight space for boosting each other. Through extensive experiments on several benchmark datasets, we demonstrated that our method achieves consistent gains in both personalization performance and robustness to problematic scenarios possible in realistic services.","url_abs":"https://arxiv.org/abs/2109.07628v3","url_pdf":"https://arxiv.org/pdf/2109.07628v3.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":"subspace-learning-for-personalized-federated","repo_url":"https://github.com/vaseline555/superfed","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"ensemble-learning","task_name":"Ensemble Learning"},{"task_slug":"federated-learning","task_name":"Federated Learning"},{"task_slug":"personalized-federated-learning","task_name":"Personalized Federated Learning"}],"methods":[{"method_slug":"orthogonal-regularization","method_name":"Orthogonal Regularization"},{"method_slug":"proximity-regularization","method_name":"Proximity Regularization"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/personalized-federated-learning-on-cifar-10","task":"Personalized Federated Learning","dataset":"CIFAR-10","model":"SuPerFed-MM","rank_in_archive_order":3,"of":7,"metrics":{"ACC@1-100Clients":"93.25","ACC@1-500Clients":"90.81","ACC@1-50Clients":"94.05"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-cifar-10","task":"Personalized Federated Learning","dataset":"CIFAR-10","model":"SuPerFed-LM","rank_in_archive_order":4,"of":7,"metrics":{"ACC@1-100Clients":"93.20","ACC@1-500Clients":"89.63","ACC@1-50Clients":"93.88"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-cifar-100","task":"Personalized Federated Learning","dataset":"CIFAR-100","model":"SuPerFed-LM","rank_in_archive_order":6,"of":7,"metrics":{"ACC@5-100Clients":"62.50"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-cifar-100","task":"Personalized Federated Learning","dataset":"CIFAR-100","model":"SuPerFed-MM","rank_in_archive_order":7,"of":7,"metrics":{"ACC@5-100Clients":"60.14"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-femnist","task":"Personalized Federated Learning","dataset":"FEMNIST","model":"SuPerFed-MM","rank_in_archive_order":1,"of":2,"metrics":{"Acc@1":"85.20","Acc@5":"99.16"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-femnist","task":"Personalized Federated Learning","dataset":"FEMNIST","model":"SuPerFed-LM","rank_in_archive_order":2,"of":2,"metrics":{"Acc@1":"83.36","Acc@5":"98.81"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-mnist-1","task":"Personalized Federated Learning","dataset":"MNIST","model":"SuPerFed-LM","rank_in_archive_order":1,"of":2,"metrics":{"ACC@1-100Clients":"99.31","ACC@1-500Clients":"98.83","ACC@1-50Clients":"99.48"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-mnist-1","task":"Personalized Federated Learning","dataset":"MNIST","model":"SuPerFed-MM","rank_in_archive_order":2,"of":2,"metrics":{"ACC@1-100Clients":"99.38","ACC@1-500Clients":"99.24"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on","task":"Personalized Federated Learning","dataset":"Shakespeare","model":"SuPerFed-MM","rank_in_archive_order":1,"of":2,"metrics":{"Acc@1":"54.52","Acc@5":"84.27"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on","task":"Personalized Federated Learning","dataset":"Shakespeare","model":"SuPerFed-LM","rank_in_archive_order":2,"of":2,"metrics":{"Acc@1":"54.52","Acc@5":"83.97"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-tiny","task":"Personalized Federated Learning","dataset":"Tiny ImageNet","model":"SuPerFed-MM","rank_in_archive_order":1,"of":2,"metrics":{"ACC@5-200Clients":"50.07"},"uses_additional_data":false},{"leaderboard":"/sota/personalized-federated-learning-on-tiny","task":"Personalized Federated Learning","dataset":"Tiny ImageNet","model":"SuPerFed-LM","rank_in_archive_order":2,"of":2,"metrics":{"ACC@5-200Clients":"49.29"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2109.07628","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2109.07628"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vaseline555/superfed","reach":null}],"summary":{"unverified":2},"by_repo_kind":{"official":{"samples":2,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":2,"samples":[{"code_sha256_prefix":"5469feaf295f1cc9","entry":"basic_evaluate","repo":"vaseline555/superfed","repo_kind":"official","path":"src/algorithm.py","file_url":"https://github.com/vaseline555/superfed/blob/HEAD/src/algorithm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5469feaf295f1cc9"}},{"code_sha256_prefix":"fa05180c9d1ab382","entry":"superfed_evaluate","repo":"vaseline555/superfed","repo_kind":"official","path":"src/algorithm.py","file_url":"https://github.com/vaseline555/superfed/blob/HEAD/src/algorithm.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fa05180c9d1ab382"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}