Federated and Distributed Learning

My work in federated and distributed learning studies how communication constraints, data heterogeneity, privacy, and robustness interact with statistical efficiency. I am interested in algorithms that retain strong guarantees while remaining practical across decentralized and non-IID data.

On the Convergence of FedAvg on Non-IID Data
X. Li*, K. Huang*, W. Yang*, S. Wang, and Z. Zhang. ICLR, 2020 (Oral).
Statistical Estimation and Online Inference via Local SGD
X. Li, J. Liang, X. Chang, and Z. Zhang. COLT, 2022.
Mitigating the Privacy-Utility Trade-off in Decentralized Federated Learning via f-Differential Privacy
X. Li*, B. Su*, C. Wang*, Q. Long, and W. J. Su. NeurIPS, 2025 (Spotlight).
Communication-Efficient Distributed SVD via Local Power Iterations
X. Li, S. Wang, K. Chen, and Z. Zhang. ICML, 2021.
FedPower: Privacy-Preserving Distributed Eigenspace Estimation
X. Guo, X. Li, X. Chang, S. Wang, and Z. Zhang. Machine Learning, 2024.
A Random Projection Approach to Personalized Federated Learning: Enhancing Communication Efficiency, Robustness, and Fairness
Y. Han, S. Lin, X. Li, and Z. Zhang (Alphabetical). Journal of Machine Learning Research, 2024.
Personalized Federated Learning towards Communication Efficiency, Robustness, and Fairness
S. Lin*, Y. Han*, X. Li, and Z. Zhang. NeurIPS, 2022.