{"author":[{"last_name":"Internò","id":"414117134","first_name":"Christian","full_name":"Internò, Christian","department":[{"_id":"10037"}]},{"last_name":"Olhofer","full_name":"Olhofer, Markus","first_name":"Markus"},{"last_name":"Jin","first_name":"Yaochu","full_name":"Jin, Yaochu"},{"id":"18302458","orcid":"0000-0002-0935-5591","first_name":"Barbara","full_name":"Hammer, Barbara","department":[{"_id":"44329888"}],"last_name":"Hammer"}],"status":"public","publication":"2024 International Joint Conference on Neural Networks (IJCNN)","conference":{"name":"2024 International Joint Conference on Neural Networks (IJCNN)","location":"Yokohama, Japan","end_date":"2024-07-05","start_date":"2024-06-30"},"abstract":[{"lang":"eng","text":"Federated learning (FL) has emerged as a groundbreaking paradigm in machine learning (ML), offering privacypreserving collaborative model training across diverse datasets. Despite its promise, FL faces significant hurdles in non-identically and independently distributed (non-IID) data scenarios, where most existing methods often struggle with data heterogeneity and lack robustness in performance. This paper introduces Federated Loss Exploration (FedLEx), an innovative approach specifically designed to tackle these challenges. FedLEx distinctively addresses the shortcomings of existing FL methods in non-IID settings by optimizing its learning behavior for scenarios in which assumptions about data heterogeneity are impractical or unknown. It employs a federated loss exploration technique, where clients contribute to a global guidance matrix by calculating gradient deviations for model parameters. This matrix serves as a strategic compass to guide clients' gradient updates in subsequent FL rounds, thereby fostering optimal parameter updates for the global model. FedLEx effectively navigates the complex loss surfaces inherent in non-IID data, enhancing knowledge transfer in an efficient manner, since only a small number of epochs and small amount of data are required to build a strong global guidance matrix that can achieve model convergence without the need for additional data sharing or data distribution statics in a large client scenario. Our extensive experiments with stateof-the art FL algorithms demonstrate significant improvements in performance, particularly under realistic non-IID conditions, thus highlighting FedLEx's potential to overcome critical barriers in diverse FL applications."}],"publication_status":"published","type_old":["conference"],"isi":"1","external_id":{"isi":["001392668205119"]},"date_updated":"2025-03-27T07:37:28Z","year":"2024","quality_controlled":"1","_id":"3001626","place":"New York","user_id":"30564","doi":"10.1109/IJCNN60899.2024.10651455","publication_identifier":{"isbn":["979-8-3503-5932-9","979-8-3503-5931-2"],"issn":["2161-4393"]},"title":"Federated Loss Exploration for Improved Convergence on Non-IID Data","citation":{"mla":"Internò, Christian, Olhofer, Markus, Jin, Yaochu, and Hammer, Barbara. “Federated Loss Exploration for Improved Convergence on Non-IID Data”. <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>. New York: Institute of Electrical and Electronics Engineers (IEEE), 2024. IEEE International Joint Conference on Neural Networks (IJCNN).","frontiers":"Internò, C., Olhofer, M., Jin, Y., and Hammer, B. (2024). “Federated Loss Exploration for Improved Convergence on Non-IID Data” in <em>2024 International Joint Conference on Neural Networks (IJCNN)</em> IEEE International Joint Conference on Neural Networks (IJCNN) (New York: Institute of Electrical and Electronics Engineers (IEEE).","chicago":"<div style=\"text-indent:-25px; padding-left:25px;padding-bottom:0px;\">Internò, Christian, Olhofer, Markus, Jin, Yaochu, and Hammer, Barbara. 2024. “Federated Loss Exploration for Improved Convergence on Non-IID Data”. In <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>. IEEE International Joint Conference on Neural Networks (IJCNN). New York: Institute of Electrical and Electronics Engineers (IEEE).</div>","angewandte-chemie":"C.  Internò, M.  Olhofer, Y.  Jin, and B.  Hammer, in <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>, Institute Of Electrical And Electronics Engineers (Ieee), New York, <strong>2024</strong>.","wels":"Internò, C.; Olhofer, M.; Jin, Y.; Hammer, B. (2024): Federated Loss Exploration for Improved Convergence on Non-IID Data. In: 2024 International Joint Conference on Neural Networks (IJCNN). New York: Institute of Electrical and Electronics Engineers (IEEE). (IEEE International Joint Conference on Neural Networks (IJCNN), ).","dgps":"<div style=\"text-indent:-25px; padding-left:25px;padding-bottom:0px;\">Internò, C., Olhofer, M., Jin, Y. &amp; Hammer, B. (2024). Federated Loss Exploration for Improved Convergence on Non-IID Data (IEEE International Joint Conference on Neural Networks (IJCNN). <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>. Gehalten auf der 2024 International Joint Conference on Neural Networks (IJCNN), New York: Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/IJCNN60899.2024.10651455.</div>","ieee":" C. Internò, et al., “Federated Loss Exploration for Improved Convergence on Non-IID Data”, <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>, IEEE International Joint Conference on Neural Networks (IJCNN),  New York: Institute of Electrical and Electronics Engineers (IEEE), 2024.","default":"Internò C, Olhofer M, Jin Y, Hammer B (2024)  <br />In:  2024 International Joint Conference on Neural Networks (IJCNN). IEEE International Joint Conference on Neural Networks (IJCNN).  New York: Institute of Electrical and Electronics Engineers (IEEE).","ama":"Internò C, Olhofer M, Jin Y, Hammer B. Federated Loss Exploration for Improved Convergence on Non-IID Data. In:  <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>. IEEE International Joint Conference on Neural Networks (IJCNN). New York: Institute of Electrical and Electronics Engineers (IEEE);  2024.","apa":"Internò, C., Olhofer, M., Jin, Y., &amp; Hammer, B. (2024). Federated Loss Exploration for Improved Convergence on Non-IID Data. <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>, IEEE International Joint Conference on Neural Networks (IJCNN) New York: Institute of Electrical and Electronics Engineers (IEEE). <a href=\"https://doi.org/10.1109/IJCNN60899.2024.10651455\" target=\"_blank\">https://doi.org/10.1109/IJCNN60899.2024.10651455</a>","apa_indent":"<div style=\"text-indent:-25px; padding-left:25px;padding-bottom:0px;\">Internò, C., Olhofer, M., Jin, Y., &amp; Hammer, B. (2024). Federated Loss Exploration for Improved Convergence on Non-IID Data. <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>, IEEE International Joint Conference on Neural Networks (IJCNN) New York: Institute of Electrical and Electronics Engineers (IEEE). <a href=\"https://doi.org/10.1109/IJCNN60899.2024.10651455\" target=\"_blank\">https://doi.org/10.1109/IJCNN60899.2024.10651455</a></div>","harvard1":"Internò, C., et al., 2024. Federated Loss Exploration for Improved Convergence on Non-IID Data. In <em>2024 International Joint Conference on Neural Networks (IJCNN)</em>. IEEE International Joint Conference on Neural Networks (IJCNN).  New York: Institute of Electrical and Electronics Engineers (IEEE).","lncs":" Internò, C., Olhofer, M., Jin, Y., Hammer, B.: Federated Loss Exploration for Improved Convergence on Non-IID Data. 2024 International Joint Conference on Neural Networks (IJCNN). IEEE International Joint Conference on Neural Networks (IJCNN).  Institute of Electrical and Electronics Engineers (IEEE), New York (2024).","bio1":"Internò C, Olhofer M, Jin Y, Hammer B (2024) <br />Federated Loss Exploration for Improved Convergence on Non-IID Data. <br />In:  2024 International Joint Conference on Neural Networks (IJCNN).  IEEE International Joint Conference on Neural Networks (IJCNN),  New York: Institute of Electrical and Electronics Engineers (IEEE)."},"department":[{"_id":"44329888"},{"_id":"10037"}],"first_contributor":"Internò, Christian","language":[{"iso":"eng"}],"unibi_person":"1","publisher":"Institute of Electrical and Electronics Engineers (IEEE)","keyword":["Federated Learning","Neural Networks","Knowledge Transfer","Non-IID data"],"series_title":"IEEE International Joint Conference on Neural Networks (IJCNN)","type":"conference","date_created":"2025-03-18T07:05:48Z"}