Peng Chen

About

I am a Ph.D. student in the Department of Computer Science at Zhejiang University, advised by Prof. Shuiguang Deng, and co-advised by Prof. Hailiang Zhao (ZJU) and Prof. Xueyan Tang (NTU). My research seeks insightful ideas in theoretical algorithm design with the development of systems. In particular, I am interested in bridging the gap between the theoretical guarantees of algorithms and practical performance in ML for Systems.

Current Research

Learning-Augmented Algorithms (a.k.a. Algorithms with Predictions) Learning-Augmented Systems

Selected Publications

[arXiv] Toward Robust and Efficient ML-Based GPU Caching for Modern Inference
Peng Chen, Jiaji Zhang, Hailiang Zhao*, Yirong Zhang, Shenyao Chen, Jiahong Yu, Xueyan Tang, Yixuan Wang, Hao Li, Jianping Zou, Gang Xiong, Kingsum Chow, Shuibing He, Shuiguang Deng*
Abstract: This paper explores ML for GPU caching systems with theoretical robustness guarantees.
[ICML'26] Towards Optimal Robustness in Learning-Augmented Paging
Peng Chen, Hailiang Zhao*, Xueyan Tang, Yixuan Wang, Shuiguang Deng*
Abstract: This paper achieves the best possible robustness for learning-augmented paging.
[NeurIPS'25] Robustifying Learning-Augmented Caching Efficiently without Compromising 1-Consistency
Peng Chen, Hailiang Zhao*, Jiaji Zhang, Xueyan Tang, Yixuan Wang, Shuiguang Deng*
Abstract: This paper improves robustness of ML-based paging algorithms while preserving 1-consistency. [arXiv version]
[NeurIPS'24] Learning-Augmented Algorithms for the Bahncard Problem
Hailiang Zhao, Xueyan Tang*, Peng Chen, and Shuiguang Deng
Abstract: This paper proposes a robust learning-augmented algorithm for the Bahncard problem. [arXiv version]

Awards

Feeds

Learning-augmented Algorithms Machine Learning for Systems Fun
Everything around us happens on this small planet called Earth

Contact

Feel free to reach out if you’d like to discuss papers or algorithms.

Email: naturechenpeng@gmail.com, pgchen@zju.edu.cn