Biography

Shuze Daniel Liu is a Research Scientist at Meta, working on LLM fine-tuning with Reinforcement Learning. His research focuses on Reinforcement Learning. He received his Ph.D. in Computer Science from the University of Virginia, advised by Professor Shangtong Zhang, his M.S. from Yale University, and his B.S. from Rensselaer Polytechnic Institute. He regularly serves on the Program Committee for major AI conferences including ICML, NeurIPS, ICLR, and AAAI.

Publications

• Doubly Optimal Policy Evaluation for Reinforcement Learning.
  Shuze Daniel Liu, Claire Chen, Shangtong Zhang.
  International Conference on Learning Representations (ICLR), 2025

• Efficient Policy Evaluation with Safety Constraint for Reinforcement Learning.
  Claire Chen*, Shuze Daniel Liu*, Shangtong Zhang.
  International Conference on Learning Representations (ICLR), 2025

• The ODE Method for Stochastic Approximation and Reinforcement Learning with Markovian Noise.
  Shuze Daniel Liu, Shuhang Chen, Shangtong Zhang.
  Journal of Machine Learning Research (JMLR), 2025.

• Efficient Multi-Policy Evaluation for Reinforcement Learning.
  Shuze Daniel Liu, Claire Chen, Shangtong Zhang.
  AAAI Conference on Artificial Intelligence (AAAI), 2025.
  Oral Presentation, 4.7%

• Efficient Policy Evaluation with Offline Data Informed Behavior Policy Design.
  Shuze Daniel Liu, Shangtong Zhang.
  International Conference on Machine Learning (ICML), 2024.

• Optimal Pricing of Information.
  Shuze Daniel Liu, Weiran Shen, Haifeng Xu.
  ACM Conference on Economics and Computation (EC), 2021.

• Strengthening Smart Contracts to Handle Unexpected Situations.
  Shuze Daniel Liu, Farhad Mohsin, Oshani Seneviratne, Lirong Xia.
  IEEE International Conference on Decentralized Applications and Infrastructures, 2019.

* indicates equal contribution

Program Committee

ICML, NeurIPS, ICLR, AAAI.

Guest Lecture

Reinforcement Learning from Human Feedback (Fall 2024),
Reinforcement Learning (Spring 2024).

Teaching Assistant

Reinforcement Learning, Machine Learning, Artificial Intelligence.