About Me
I am a Post Doctoral Scholar at the School of Biomedical Informatics, Ohio State University. In September 2024, I completed my Ph.D. degree at the School of Computer Science and Technology, Xi’an Jiaotong University, advised by Prof. Chen Li and Prof. Tieliang Gong. I obtained my B.E. degree in Computer Science and Technology at Xi’an Jiaotong University.
My research interests lie in machine learning and statistical learning theory. Recently, I have been focusing on information-theoretic generalization analysis and robust learning in areas of supervised learning, contrastive learning, and domain generalization. These works shed light on understanding the success and limitations of existing algorithms or inspire new algorithm designs that are provably more effective. My main research topics include:
- Analyzing the generalization ability of randomized learning algorithms through the lens of information theory.
- Designing effective and robust learning algorithms based on information-theoretic measurements and analysis.
- Developing computationally efficient approximations for information-theoretic quantities and measurements.
Selected Publications
How Does Distribution Matching Help Domain Generalization: An Information-theoretic Analysis.
Yuxin Dong, Tieliang Gong, Hong Chen, Shuangyong Song, Weizhan Zhang, Chen Li. IEEE Transactions on Information Theory. 2025. [Link] [PDF]
Towards Generalization beyond Pointwise Learning: A Unified Information-theoretic Perspective.
Yuxin Dong, Tieliang Gong, Hong Chen, Mengxiang Li, Zhongjiang He, Shuangyong Song, Chen Li. International Conference on Machine Learning. 2024. [Link] [PDF] [Code]
Rethinking Information-theoretic Generalization: Loss Entropy Induced PAC Bounds.
Yuxin Dong, Tieliang Gong, Hong Chen, Shujian Yu, Chen Li. International Conference on Learning Representations. 2024. [Link] [PDF] [Code]
Optimal Randomized Approximations for Matrix-based Rényi's Entropy.
Yuxin Dong, Tieliang Gong, Shujian Yu, Chen Li. IEEE Transactions on Information Theory. 2023. [Link] [PDF]
Understanding the Generalization Ability of Deep Learning Algorithms: A Kernelized Rényi’s Entropy Perspective.
Yuxin Dong, Tieliang Gong, Hong Chen, Chen Li. International Joint Conference on Artificial Intelligence. 2023. [Link] [PDF] [Code]
Robust and Fast Measure of Information via Low-Rank Representation.
Yuxin Dong, Tieliang Gong, Shujian Yu, Hong Chen, Chen Li. AAAI Conference on Artificial Intelligence. 2023. [Link] [PDF] [Code]
Efficient Approximations for Matrix-Based Rényi’s Entropy on Sequential Data.
Yuxin Dong, Tieliang Gong, Hong Chen, Chen Li. IEEE Transactions on Neural Networks and Learning Systems. 2023. [Link] [PDF]
PhD Thesis
Information-theoretic Generalization Analysis for Randomized Learning Algorithms.
Yuxin Dong. PhD Thesis. 2024. [Link] [PDF]