Research

My research interests lie broadly in sequential decision-making under uncertainty, including online learning, reinforcement learning, and resource allocation, with applications to experimentation, supply chain management and revenue management.

• Methodologically, beyond merely focusing on maximizing short-term efficiency, my research delves into the safety, robustness, and resilience of AI-assisted decision-making systems, promoting long-term efficiency through managing hidden risk:
  • enhancing policy safety and robustness during and after online experiments [1, 5, 6] — our results provide theoretical insights to the practical success of AlphaGo MCTS;
  • risk detection, response coordination, and system recovery for supply chain disruptions [3, 7] — our results showcase the approriate approach of handling uncertain disruption predictions.

• Practically, I have strong interests in applying my research philosophy into supply chains and finance.
  • I have been working with several large companies (Accenture, DENSO, Ford) to help design and implement supply chain risk detection models and mitigation strategies. In Summer 2023, I spent a great time working on inventory simulation and optimization as a data science intern at Ford.
  • In Summer 2025, I spent a great time working on developing trading strategies as a quantitative research intern at Citadel LLC.

Publications, Preprints, Articles (*represents α-β order)