Research
I build data-driven decision tools for retail and supply-chain, with a focus on problems that are operationally important, mathematically hard, and easy to get wrong in practice. What excites me is turning messy, high-dimensional reality into methods that are simultaneously simple enough to implement, robust enough to trust, and rigorous enough to prove something meaningful about. I’m especially drawn to questions where naïve optimization breaks down and where the right model/algorithm choice can produce large gains in efficiency, service, and profitability. Much of my work is motivated by real collaborations, and I’m always open to partnerships where theory and deployment can meet
Selected Publications & Working Papers
-
Return Concerned Adaptive Control of Product Offerings
Paper -
Online Learning of Delayed Choices
Neural Information Processing Systems (NeurIPS), 2024
Paper -
Inventory Control and Learning for One-Warehouse Multistore System with
Censored Demand
Operations Research, 2023
Paper -
Probabilistic Models for Manufacturing Lead Times
arXiv -
Visualizing the Loss Landscape of Actor-Critic Methods with Applications
in Inventory Optimization
arXiv
Patents
-
Machine learning based federated learning with hierarchical modeling hotel
upsell
-
Multi-Product Inventory Assortment and Allocation Optimization