Transaction

jmetaB026d8b4e53b04950dd08ba56b04603284ee87507cd719b88135b1f1bb08f3cc48c@104e04f4dc7bbb58b675a0be8ec8a2392cd828cadc0c1b85347e2d4ab003150erss.itemmetarss.netM`<item> <title>Accuracy-Privacy Trade-off in the Mitigation of Membership Inference Attack in Federated Learning</title> <link>https://arxiv.org/abs/2407.19119</link> <description>arXiv:2407.19119v1 Announce Type: cross Abstract: Over the last few years, federated learning (FL) has emerged as a prominent method in machine learning, emphasizing privacy preservation by allowing multiple clients to collaboratively build a model while keeping their training data private. Despite this focus on privacy, FL models are susceptible to various attacks, including membership inference attacks (MIAs), posing a serious threat to data confidentiality. In a recent study, Rezaei \textit{et al.} revealed the existence of an accuracy-privacy trade-off in deep ensembles and proposed a few fusion strategies to overcome it. In this paper, we aim to explore the relationship between deep ensembles and FL. Specifically, we investigate whether confidence-based metrics derived from deep ensembles apply to FL and whether there is a trade-off between accuracy and privacy in FL with respect to MIA. Empirical investigations illustrate a lack of a non-monotonic correlation between the number of clients and the accuracy-privacy trade-off. By experimenting with different numbers of federated clients, datasets, and confidence-metric-based fusion strategies, we identify and analytically justify the clear existence of the accuracy-privacy trade-off.</description> <guid isPermaLink="false">oai:arXiv.org:2407.19119v1</guid> <category>cs.LG</category> <category>cs.AI</category> <category>cs.CR</category> <arxiv:announce_type>cross</arxiv:announce_type> <dc:rights>http://creativecommons.org/licenses/by-nc-nd/4.0/</dc:rights> <dc:creator>Sayyed Farid Ahamed, Soumya Banerjee, Sandip Roy, Devin Quinn, Marc Vucovich, Kevin Choi, Abdul Rahman, Alison Hu, Edward Bowen, Sachin Shetty</dc:creator> </item>