Due to its communication efficiency and privacy-preserving capability, federated learning has emerged as a promising framework for machine learning in 5G-and-beyond wireless networks. Therefore, the novel designs of wireless networks to support the stable and fast operation of federated learning are of great interest. This talk will first highlight the reasons why federated learning becomes important in the future, briefly introduce communication- oriented research directions for realizing federated learning in wireless networks and then discuss state-of-the-art solutions for a massive multiple-input multiple-output (MIMO) network to efficiently support federated learning.
Speaker: TS. Vũ Thanh Tùng, Queen’s Univ. Belfast
Time: 15:30, Tuesday, March 29, 2022
Venue: Webinar
Thanh Tung Vu is currently a Research Fellow at Queen’s University Belfast, UK. His current research interests include applied optimization and machine learning theories in future wireless networks (e.g., massive MIMO, distributed machine learning systems, next-generation satellites). He was the author of the first research work using Cell-free Massive MIMO to support Federated Learning (Top 50 popular papers in Oct-Dec. 2020 of IEEE Transactions on Wireless Communications). He is currently serving as an Editor of Elsevier Physical Communication (PHYCOM). He was an Exemplary Reviewer for IEEE Transactions on Communications in 2021 and IEEE Wireless Communication Letters in 2020 and 2021.