dc.contributor.author | Odeyomi, Olusola T. | |
dc.contributor.author | Záruba, Gergely V. | |
dc.date.accessioned | 2021-10-18T02:17:19Z | |
dc.date.available | 2021-10-18T02:17:19Z | |
dc.date.issued | 2021-07-12 | |
dc.identifier.citation | Odeyomi, O., & Zaruba, G. (2021). Differentially-private federated learning with long-term constraints using online mirror descent. Paper presented at the IEEE International Symposium on Information Theory - Proceedings, , 2021-July 1308-1313. doi:10.1109/ISIT45174.2021.9518177 | en_US |
dc.identifier.isbn | 978-1-5386-8209-8 | |
dc.identifier.isbn | 978-1-5386-8210-4 | |
dc.identifier.uri | https://doi.org/10.1109/ISIT45174.2021.9518177 | |
dc.identifier.uri | https://soar.wichita.edu/handle/10057/22209 | |
dc.description | Click on the DOI link to access this conference paper at the publishers website (may not be free). | en_US |
dc.description.abstract | This paper discusses a fully decentralized online federated learning setting with long-term constraints. The fully decentralized setting removes communication and computational bottlenecks associated with a central server communicating with a large number of clients. Also, online learning is introduced to the federated learning setting to capture situations with time-varying data distribution. Practical federated learning settings are imposed with long-term constraints such as energy constraints, money cost constraints, time constraints etc. The clients are not obligated to satisfy any per round constraint, but they must satisfy these long-term constraints. To provide privacy to the shared local model updates of the clients, local differential privacy is introduced. An online mirror descent-based algorithm is proposed and its regret bound is obtained. The regret bound is compared with the regret bound of a differentially-private version of online gradient descent algorithm proposed for federated learning. | en_US |
dc.language.iso | en_US | en_US |
dc.publisher | IEEE | en_US |
dc.relation.ispartofseries | 2021 IEEE International Symposium on Information Theory (ISIT); | |
dc.subject | Differential privacy | en_US |
dc.subject | Privacy | en_US |
dc.subject | Computational modeling | en_US |
dc.subject | Aggregates | en_US |
dc.subject | Training data | en_US |
dc.subject | Collaborative work | en_US |
dc.subject | Servers | en_US |
dc.title | Differentially-private federated learning with long-term constraints using online mirror descent | en_US |
dc.type | Conference paper | en_US |
dc.rights.holder | Copyright © 2021, IEEE | en_US |