Federated Learning
Distributed and federated learning algorithms for heterogeneous data
Federated learning enables multiple agents to collaboratively train a shared model while keeping their data decentralized. Our group studies fundamental challenges in federated learning including communication efficiency, non-IID data, privacy, and personalization.
Selected Publications
- Xinwei Zhang, Siliang Zeng, et al., “FedPD: A Federated Learning Framework with Optimal Rates and Adaptivity to Non-IID Data”
- Tianyi Chen, Xinwei Zhang, Wotao Yin, Mingyi Hong, “Hybrid Federated Learning: Algorithms and Implementation”, NeurIPS 2020 Workshop on Scalability, Privacy, and Security in Federated Learning. Best Student Paper Award