2021 2nd International Conference on Mobile Computing, Wireless Communications and Networks
Prof. Jun Li

Prof. Jun Li

李骏 116x160.jpg

Prof. Jun Li

Nanjing University of Science and Technology, China

>>>Personal Webpage: Click

Brief Introduction:

Jun Li is a Professor at the School of Electronic and Optical Engineering, Nanjing University of Science and Technology, Nanjing, China. He was a visiting professor at Princeton University from 2018 to 2019. His research interests include network information theory, game theory, distributed intelligence, multiple agent reinforcement learning, and their applications in ultra-dense wireless networks, mobile edge computing, network privacy and security, and industrial Internet of things. He has co-authored more than 200 papers in IEEE journals and conferences and holds 1 US patent and more than 10 Chinese patents in these areas. He is serving as an associate editor of IEEE Transactions on Wireless Communications, a guest editor of IEEE Journal of Selected Topics in Signal Processing, and TPC member for several flagship IEEE conferences.

Speech title: 

Privacy and Security for Federated Learning: Performance Analysis and Mechanism Design


Motivated by the advancing computational capacity of wireless end-user equipment (UE), as well as the increasing concerns about sharing private data, a new machine learning (ML) paradigm has emerged, namely federated learning (FL). Specifically, FL allows decoupling of data provision at UEs and ML model aggregation at a central unit. By training the model locally, FL is capable of avoiding direct data leakage from the UEs, thereby preserving privacy and security to some extent. However, even if raw data are not disclosed from UEs, individual's private information can still be extracted by some recently discovered attacks against the FL architecture. In this talk, we mainly provide three attractive sections to analyze the privacy and security issues in FL and discuss several challenges on preserving privacy and security when designing FL systems. In detail, for the privacy issue, we propose a differentially private FL framework by adding appropriate noises on the parameters and analyzing the convergence performance in terms of the privacy level. Then for the security issue, we propose two potential attacking methods on the current FL framework, and the system performance and defensive mechanisms are investigated in this part. In addition, to avoid the one-point-failure issue existing in the FL framework, we propose a blockchain-aided FL framework. The procedure details and corresponding system performance are also included in this section.