Scholar
Wenzhe Li
Google Scholar ID: WMC5PsAAAAAJ
Princeton University
Citations & Impact
Citations
497
H-index
9
i10-index
9
Publications
11
Co-authors
15
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Published papers at ICML 2024, ICLR AGI Workshop 2024 (FightLadder: A Benchmark for Competitive Multi-Agent Reinforcement Learning); AAAI 2023 (Flow to Control: Offline Reinforcement Learning with Lossless Primitive Discovery); NeurIPS 2022 (Latent-Variable Advantage-Weighted Policy Optimization for Offline RL), etc. Ranked 1st on AlpacaEval 2.0 Leaderboard (non-adversarial).
Research Experience
- Teaching Assistant for Deep Reinforcement Learning (Tsinghua University, Mar 2022); Teaching Assistant for Artificial Intelligence: Principles and Techniques (Tsinghua University, Sep 2021). Participated in multiple research projects, including the design of a benchmark platform for competitive multi-agent reinforcement learning called FightLadder.
Education
- Ph.D. in Electrical and Computer Engineering, Princeton University, 2023 -- Present, Advisor: Chi Jin; B.E. in Computer Science and Technology, Tsinghua University, 2017 -- 2021, Advisors: Chongjie Zhang, Guy Van den Broeck, Stefano Ermon
Background
- Research interests include LLM reasoning & agents, reinforcement learning, and multi-agent learning. Has worked on topics such as large language models, multi-agent reinforcement learning, offline reinforcement learning, score-based generative models, and approximate inference.
Miscellany
- Email: scott.wenzhe.li at gmail.com
