Basic Information:
Dr Ruizhi Pu is an Assistant Professor at the School of Statistics and Data Science, Southeast University. He studied at Western University in Canada from September 2021 to September 2025 and obtained a Doctoral Degree in Computer Science. He participated in multiple research internships at research institutions such as Tencent AI Lab (2024/2025) and Huawei Noah's Ark Lab (2022/2025).
Research area:
His main research areas include robust machine learning (label noise, data imbalance), federated learning, and graph neural network learning.
Selected Papers:
[1]Graph Domain Adaptation via Homophily-Agnostic Reconstructing Structure. Ruiyi Fang,…, Ruizhi Pu, et al. AAAI-2026, Oral.
[2]Leveraging group classification with descending soft labeling for deep imbalanced regression. Ruizhi Pu, et al. AAAI-2025, Oral.
[3]On the benefits of attribute-driven graph domain adaptation. Ruiyi Fang, .., Ruizhi Pu, et al. ICLR-2025.
[4]Towards more general loss and setting in unsupervised domain adaptation. CJ Shui*, Ruizhi Pu*，et al. TKDE.
[5]FedELR: When federated learning meets learning with noisy labels. Ruizhi Pu, et al. Neural Networks.
[6]FedFMD: Fairness-Driven Adaptive Aggregation in Federated Learning via Mahalanobis Distance. XiuTing Wen, …, Ruizhi Pu, et al. CIKM 2025.
[7]Unraveling the mysteries of label noise in source-free domain adaptation: Theory and practice. Gezheng Xu,…,Ruizhi Pu, et al. TPAMI.
[8]When source-free domain adaptation meets learning with noisy labels. Yi Li,…, Ruizhi Pu, et al. ICLR-2023 Spotlight.
