📖 Brief Bio
Hi, I am currently a Ph.D. student in Computing and Information Sciences at Golisano College of Computing and Information Sciences at Rochester Institute of Technology, advised by Prof. Haibo Yang. My research interests mainly concentrate on addressing issues caused by heterogeneity in Federated Learning and improving communication and memory efficiency of federated or large-scale distributed systems by optimization techniques (e.g., zeroth-order optimization).
My research interests include:
- Distributed Machine Learning/Federated Learning
- Optimization Theory
🗞️ News
- [2025.01] Our paper “Achieving Dimension-Free Communication in Federated Learning via Zeroth-Order Optimization” has been accepted to ICLR 2025.
- [2024.10] Our paper “DeComFL: Federated Learning with Dimension-Free Communication” has been accepted to FL@FM-NeurIPS 2024 (oral).
- [2024.08] Joined Rochester Institute of Technology(RIT) for my Ph.D. in Computing and Information Sciences, advised by Prof. Haibo Yang, and started to serve as a Graduate Research Assistant at RIT.
- [2024.05] Successfully obtained my master’s degree from the University of Southern California. Fight On!
📝 Publications
- Zhe Li, Bicheng Ying, Zidong Liu, Chaosheng Dong, and Haibo Yang. Achieving Dimension-Free Communication in Federated Learning via Zeroth-Order Optimization. International Conference on Learning Representations, 2025. [Openreview] [Arxiv] [Code]
- Zhe Li, Bicheng Ying, Zidong Liu, Chaosheng Dong, and Haibo Yang. DeComFL: Dimension-Free Communication in Federated Learning. International Workshop on Federated Foundation Models in Conjunction with NeurIPS, 2024. [Paper] [Code]
🎓 Education
- 2024.08 - 2029.05, Ph.D. in Computing and Information Sciences,
Rochester Institute of Technology, Rochester, New York, USA - 2022.08 - 2024.05, M.S. in Electrical Engineering,
University of Southern California, Los Angeles, California, USA - 2015.09 - 2019.07, B.E. in Software Engineering,
Shanghai University of Electric Power, Shanghai, CHN
🏛️ Service
- Reviewer: KDD 2024, INFOCOM 2025
