Research Interests
1. Self-Adapting Approximate Dynamic Programming: My research advances frameworks for dynamic decision making under uncertainty that are accessible to non-expert users. Rather than requiring users to manually design models, tune parameters, and solve large-scale optimization problems through trial and error, these frameworks automatically design, compute, and enhance their own approximations and produce high-quality decisions. By self-adapting to each problem instance's data and using information revealed during computation, these methods substantially reduce deployment complexity while maintaining strong theoretical performance guarantees, making advanced decision-support tools more accessible to users with limited domain expertise.
2. AI-Optimization Hybrids for Modern Business Applications: My research investigates emerging operations management problems shaped by real-world operational complexities, including operating-room scheduling systems that account for evolving workforce composition, pricing and demand-learning mechanisms that maintain consistency across retail channels, and marketing campaigns that account for long-term and cross-store effects. To address these challenges, I develop tailored algorithms that integrate AI and data-driven optimization technologies to deliver high-quality decision support and generate managerial insights for these complex dynamic systems.
For a complete list of my published and working papers, please visit my research page.