Topic: Learning Customer Preferences from Personalized Assortments

Speaker: Mr. Yifan Feng, University of Chicago

Time & Date: 16:00 - 17:00, Tuesday, December 10, 2019

Venue: Room 208, Cheng Dao Building

Abstract：

A company wishes to commercialize a single version of a product from a menu of alternative options. Unaware of true customer preferences, the company relies on a system that allows potential buyers to provide feedback on their preferred versions. Under a general ranking-based choice model framework, we study how to dynamically individualize the set of versions shown to each customer on which they provide feedback. We prove an instance-specific lower bound on the sample complexity of any policy that identifies the top-ranked version with a given (probabilistic) confidence. We also propose a robust formulation of the company's problem and derive a sampling policy (Myopic Tracking Policy), which is both asymptotically sample optimal and intuitive to implement. We conduct computational studies on both synthetic and real-life data to assess the performance of our proposed method and compare it to alternative methods proposed in the literature.

Biography：

Yifan Feng is a Ph.D. Candidate at the University of Chicago Booth School of Business. He is broadly interested in dynamic learning and data-driven modeling in online platform operations and market analytics. His research employs tools in statistical machine learning, economics, and empirical methods to address operational issues such as preference learning, pricing, and demand fulfillment. Methodologically, his contributions belong to the areas of sequential design of experiments, as well as learning in presence of strategic behavior. He actively collaborates with industry partners such as the Alibaba Group, among others. Before Booth, Yifan received a bachelor's degree (summa cum laude) in mathematics from the University of Minnesota.