On experimentation with heterogeneous subgroups: An asymptotic optimal δ-weighted-pac design
David Simchi-Levi, Chonghuan Wang, Jiamin Xu (authors listed in alphabetical order)
CV
Education
- Ph.D in Operations Research, Cornell Tech, 2029 (expected)
- B.S. in Mathematics, Fudan University, 2024
Skills
- Pytorch
- Reinforcement Learning
- Deep RL
- Causal RL
- Bandit Algorithms
Publications
From Restless to Contextual: A Thresholding Bandit Reformulation For Finite-horizon Performance
Jiamin Xu, Ivan Nazarov, Aditya Rastogi, África Periáñez, Kyra Gan
AISTATS 2026
Fast Non-Episodic Finite-Horizon RL with K-Step Lookahead Thresholding
Jiamin Xu, Kyra Gan
ICML 2026
Integrating Causal DAGs in Deep RL: Activating Minimal Markovian States with Multi-Order Exposure
Jiamin Xu, Jacqueline Maasch, Kyra Gan
ICML 2026 SPIGM Workshop
Talks
Fast Non-Episodic Finite-Horizon RL with K-Step Lookahead Thresholding
Talk at INFORMS Healthcare Conference 2026, Raleigh, NC, USA
Integrating Causal DAGs in Deep RL: Activating Minimal Markovian States with Multi-Order Exposure
Talk at Amazon Scout RL Lunch Talk, New York, NY, USA
From Restless to Contextual: A Thresholding Bandit Reformulation For Finite-horizon Performance
Talk at INFORMS Annual Meeting 2025, Atlanta, GA, USA
Service and leadership
- Co-chairing one INFORMS Annual Meeting 2025 Session