Chung-Wei Lee

[Google Scholar]
[email protected]

I am a research scientist at Meta. Prior to that, I spent my time at WorldQuant conducting machine learning and artificial intelligence research in quantitative finance. I hold a PhD in Computer Science at the University of Southern California, where I was very fortunate to be advised by Prof. Haipeng Luo. My PhD research centers around the intersection of theoretical machine learning and algorithmic game theory. Specifically, I study fundamental sequential decision-making problems involving multiple agents in partially observable environments. Prior to pursuing my PhD, I received my B.S. from National Taiwan University, double majoring in Electrical Engineering and Mathematics. During that time, I had the privilege of working alongside Prof. Yu-Chiang Frank Wang on Computer Vision and Deep Learning projects.


Work Experience


10/2024 - present

Research Scientist at Meta in Sunnyvale

01/2024 - 09/2024

Researcher at WorldQuant in Taipei


Internship Experience


06/2023 - 07/2023

Research Intern at WorldQuant in Taipei

09/2022 - 01/2023

Research Intern at DeepMind in London

05/2022 - 08/2022

Research Intern at Google Research in Mountain View

01/2022 - 05/2022

Research Intern at Meta AI in Menlo Park

05/2021 - 08/2021

Research Intern at ByteDance in Mountain View


Publications

Context-lumpable Stochastic Bandits
Chung-Wei Lee, Qinghua Liu, Yasin Abbasi-Yadkori, Chi Jin, Tor Lattimore, Csaba Szepesvári
Conference on Neural Information Processing Systems (NeurIPS) 2023.
[arxiv]

Regret Matching+: (In)Stability and Fast Convergence in Games
(α-β order) Gabriele Farina, Julien Grand-Clément, Christian Kroer, Chung-Wei Lee, Haipeng Luo
Conference on Neural Information Processing Systems (NeurIPS) 2023. (Spotlight)
[arxiv]

Uncoupled Learning Dynamics with O(log T) Swap Regret in Multiplayer Games
(α-β order) Ioannis Anagnostides, Gabriele Farina, Christian Kroer, Chung-Wei Lee, Haipeng Luo, Tuomas Sandholm
Conference on Neural Information Processing Systems (NeurIPS) 2022. (Oral)
[arxiv]

Near-Optimal No-Regret Learning Dynamics for General Convex Games
Gabriele Farina, Ioannis Anagnostides, Haipeng Luo, Chung-Wei Lee, Christian Kroer, Tuomas Sandholm
Conference on Neural Information Processing Systems (NeurIPS) 2022.
[arxiv]

Kernelized Multiplicative Weights for 0/1-Polyhedral Games: Bridging the Gap Between Learning in Extensive-Form and Normal-Form Games
Gabriele Farina, Chung-Wei Lee, Haipeng Luo, Christian Kroer
International Conference on Machine Learning (ICML) 2022.
[arxiv]

Last-iterate Convergence in Extensive-form Games
Chung-Wei Lee, Christian Kroer, Haipeng Luo
Conference on Neural Information Processing Systems (NeurIPS) 2021.
[arxiv]

Policy Optimization in Adversarial MDPs: Improved Exploration via Dilated Bonuses
Haipeng Luo, Chen-Yu Wei, Chung-Wei Lee
Conference on Neural Information Processing Systems (NeurIPS) 2021.
[arxiv]

Achieving Near Instance-Optimality and Minimax-Optimality in Stochastic and Adversarial Linear Bandits Simultaneously
(α-β order) Chung-Wei Lee, Haipeng Luo, Chen-Yu Wei, Mengxiao Zhang, Xiaojin Zhang
International Conference on Machine Learning (ICML) 2021.
[arxiv]

Last-iterate Convergence of Decentralized Optimistic Gradient Descent/Ascent in Infinite-horizon Competitive Markov Games
Chen-Yu Wei, Chung-Wei Lee, Mengxiao Zhang, Haipeng Luo
Annual Conference on Learning Theory (COLT) 2021.
[arxiv]

Linear Last-iterate Convergence in Constrained Saddle-point Optimization
Chen-Yu Wei, Chung-Wei Lee, Mengxiao Zhang, Haipeng Luo
Conference on Learning Representations (ICLR) 2021.
[arxiv]

Bias No More: High-probability Data-dependent Regret Bounds for Adversarial Bandits and MDPs
(α-β order) Chung-Wei Lee, Haipeng Luo, Chen-Yu Wei, Mengxiao Zhang
Conference on Neural Information Processing Systems (NeurIPS) 2020. (Oral)
[arxiv]


A Closer Look at Small-loss Bounds for Bandits with Graph Feedback
(α-β order) Chung-Wei Lee, Haipeng Luo, Mengxiao Zhang
Annual Conference on Learning Theory (COLT) 2020.
[arxiv]


A New Algorithm for Non-stationary Contextual Bandits: Efficient, Optimal, and Parameter-free
(α-β order) Yifang Chen, Chung-Wei Lee, Haipeng Luo, Chen-Yu Wei
Annual Conference on Learning Theory (COLT) 2019.
[arxiv]

Multi-Label Zero-Shot Learning with Structured Knowledge Graphs
Chung-Wei Lee, Wei Fang, Chih-Kuan Yeh, Yu-Chiang Frank Wang
IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2018.
[arxiv]

Preprints

Practical Knowledge Distillation: Using DNNs to Beat DNNs
Chung-Wei Lee, Pavlos Athanasios Apostolopulos, Igor L. Markov
Intern project report.
[arxiv]