主题：Optimizing Model Utility for Differentially Private Federated Learning

主讲嘉宾：Yipeng Zhou, Macquarie University

时间：2023年4月14日 09:00-10:00

地点：致腾楼623会议室

主持人：崔来中，教授

报告人简介：

Dr. Yipeng Zhou is a senior lecturer with School of Computing, Faculty of Science and Engineering at Macquarie University, Australia. He is the recipient of ARC Discover Early Career Researcher Award (DECRA) in 2018. He got his Ph.D. degree from Information Engineering Department of CUHK and Bachelor degree from Department of Computer Science and Technology of University of Science and Technology of China (USTC). His research interests lie in federated learning, privacy protection and  networking. He has published more than 100 papers including IEEE INFOCOM, ICNP, IWQoS, IEEE ToN, TDSC, JSAC, TPDS, TMC, TMM, etc.

报告摘要：

Federated learning (FL) empowers distributed clients to collaboratively train a shared machine learning model through exchanging parameter information. Despite the fact that FL can protect clients' raw data, malicious users can still crack original data with disclosed parameters. To amend this flaw, differential privacy (DP) is incorporated into FL clients to disturb original parameters, which however can significantly impair model utility. This talk will focus on the study how to minimize the influence of DP noises on model utility by optimizing critical hyperparameters in DP enhanced FL. We investigate how these hyperparameters including the number of participants per round and the number of conducted global iterations affect the scale of DP noises, and hence the influence on model utility. How these hyperparameters can be optimally determined will be explored as well. In the end, we envision future work in this topic.