Research

My research centers on three questions:
(1) How can we make reinforcement learning reliable and safe in high-stakes domains?
(2) How can we develop and use simulators to improve decision making?
(3) How can we integrate human expertise into data-driven decision making?

Published as Jason Meng and Jie Meng.

* Equal contribution.

Publications

TMLR’26
PERRY: Policy Evaluation with Confidence Intervals using Auxiliary Data
Aishwarya Mandyam*, Jason Meng*, Ge Gao, Jiankai Sun, Mac Schwager, Barbara Engelhardt, Emma Brunskill
IEEE TIT’24
Efficient Reinforcement Learning with Impaired Observability: Learning to Act with Delayed and Missing State Observations
Minshuo Chen, Jie Meng, Yu Bai, Yinyu Ye, H. Vincent Poor, Mengdi Wang

Manuscripts under review

2026
Trustmix: Reward Learning from Aggregate Feedback with Heterogeneous Surrogates
Haozhan Gao*, Jason Meng*
2025
Learning Optimal Distributionally Robust Stochastic Control in Continuous State Spaces
Shengbo Wang, Jason Meng, Nian Si, Jose Blanchet, Zhengyuan Zhou
2024
Optimal Policy Learning in Nonstationary Markov Decision Processes with Change Detection
Liyan Xie, Jie Meng, Minshuo Chen, H. Vincent Poor, Mengdi Wang

Conference abstracts & presentations

AHA’26
Beyond Nursing Workload: Developing an Acuity Score for the Cardiovascular Intensive Care Unit at the Stanford Children’s Hospital
Jason Meng*, Johnathan Le*, Kaitlyn Kim*, Hannah Thomas*, Aman Dhillon*, Brian Rigby*, Andrew Shin, David Scheinker, Meghna Patel, Rebecca Kameny

Research

Publications

PERRY: Policy Evaluation with Confidence Intervals using Auxiliary Data

Efficient Reinforcement Learning with Impaired Observability: Learning to Act with Delayed and Missing State Observations

Manuscripts under review

Trustmix: Reward Learning from Aggregate Feedback with Heterogeneous Surrogates

Learning Optimal Distributionally Robust Stochastic Control in Continuous State Spaces

Optimal Policy Learning in Nonstationary Markov Decision Processes with Change Detection

Conference abstracts & presentations

Beyond Nursing Workload: Developing an Acuity Score for the Cardiovascular Intensive Care Unit at the Stanford Children’s Hospital