|
Dun Zeng (曾趸)
Hi there! I am a final-year Ph.D. candidate at the University of Electronic
Science and
Technology of China (UESTC). I am very fortunately supervised by Prof.
Zenglin
Xu. Before that, I received my Bachelor’s Degree from Ocean
University of China
(OUC) in 2020.
My research interests primarily focus on (Distributed) Optimization and Generalization of (non-convex) machine learning problems (Yes, neural networks). I work on solving the challenge of conflicting optimization objectives in machine learning models, improving the efficiency of data application in human society. My Ph.D projects focused on designing, analyzing, and evaluating federated optimization algorithms, considering communication and heterogeneity to achieve a high generalization model in federated learning. Recently, I turned to studying the Large Language Models alignment, particularly interested in the RLHF framework and reward models.
I am pursuing academic collaborations, particularly in the field of distributed optimization and LLMs. Please feel free to reach out!
Email /
Scholar /
Github /
Zhihu
|
|
Publications (* denotes equal contribution)
|
-
FedLab: A Flexible Federated Learning Framework [Code]
Dun Zeng*, Siqi Liang*, Xiangjin Hu, Hui Wang, Zenglin Xu
JMLR, 2023.
-
On Diversified Preferences of Large Language Model Alignment [Code]
Dun Zeng*, Yong Dai*, Pengyu Cheng*, Longyue Wang, Tianhao Hu, Wanshun Chen, Nan Du, Zenglin Xu
EMNLP, 2024.
-
Stochastic Clustered Federated Learning [Code]
Dun Zeng, Xiangjing Hu, Shiyu Liu, Yue Yu, Qifan Wang, Zenglin Xu
KDD FL4Data-Mining Workshop, 2023.
-
On the Power of Adaptive Weighted Aggregation in Heterogeneous Federated Learning and Beyond [Code]
Dun Zeng, Zenglin Xu, Yu Pan, Qifan Wang, Xiaoying Tang
preprint, 2023.
-
Understanding Generalization of Federated Learning via Trade-offs in Model Stability and Optimization Bounds
Dun Zeng*, Zheshun Wu*, Shiyu Liu, Yu Pan, Xiaoying Tang, Zenglin Xu
preprint, 2024.
-
Enhanced Federated Optimization: Adaptive Unbiased Client Sampling with Reduced Variance [Code]
Dun Zeng, Zenglin Xu, Yu Pan, Xu Luo, Qifan Wang, Xiaoying Tang
preprint, 2023.
-
Federated Knowledge Graph Completion via Latent Embedding Sharing and Tensor Factorization
Maolin Wang, Dun Zeng, Zenglin Xu, Ruocheng Guo, Xiangyu Zhao
ICDM, 2023.
-
A Survey of Trustworthy Federated Learning with Perspectives on Security, Robustness and Privacy
Yifei Zhang*, Dun Zeng*, Jinglong Luo*, Zenglin Xu, Irwin King
WWW, 2023.
-
Topology Learning for Heterogeneous Decentralized Federated Learning over Unreliable D2D Networks
Zheshun Wu, Zenglin Xu, Dun Zeng, Junfan Li, Jie Liu
IEEE Transactions on Vehicular Technology, 2024.
|