My research focuses on data-driven and dynamic optimization and its applications in market design. I combine tools from online and stochastic optimization, machine learning, and game theory to improve the operations of modern marketplaces, including organ donation schemes, ride-sharing platforms, and emergency response systems. More recently, I have become interested in markets powered by AI agents; how such markets operate, how they can be optimized, and what AI agents are capable of in operational decision-making.
Papers
"Near-optimal Adaptive Policies for Dynamic Matching" with Ali Aouad and Amin Saberi, Operations Research (major revision)
[arXiv][video]
Appeared in the 57th Annual ACM Symposium on Theory of Computing (STOC'25).
"Spatial Matching under Multihoming" with Ali Aouad and Daniel Freund, Operations Research (minor revision)
[SSRN][poster]
"Improved Approximations for Stationary Bipartite Matching: Beyond Probabilistic Independence" with Ali Aouad, Tristan Pollner, and Amin Saberi, submitted
[arXiv]
Accepted in the 27th ACM Conference on Economics and Computation (EC'26).
"Long-term Optimization subject to Approachability" with Neil Walton, working paper.
"Stationary Bipartite Matching with Stochastic Rewards" with Ali Aouad and Amin Saberi, working paper.
"Governance of Social Welfare in Networked Markets" with Mohammadamin Fazli, IEEE Transactions on Computational Social Systems
[published]
"On the maximum order of induced paths and induced forests in regular graphs" with Saeeid Akbari, Sepehr Mousavi, Hessam Nikpey, Soheil Sheybani
[arXiv]
Work in progress
"Learning a Choice Model for Bundles via Neural Networks" with Ali Aouad and Antoine Desir.
"Bayesian Learning in Online Decision-making" with Ali Aouad, Senem Isik, and Amin Saberi.