MSIS Seminar

02 Dec
-
2022-10-07T16:00:00 2022-10-14T16:00:00 2022-10-21T16:00:00 2022-10-28T16:00:00 2022-11-04T16:00:00 2022-11-11T16:00:00 2022-11-18T16:00:00 2022-12-02T16:00:00 2022-10-07T17:00:00 2022-10-14T17:00:00 2022-10-21T17:00:00 2022-10-28T17:00:00 2022-11-04T17:00:00 2022-11-11T17:00:00 2022-11-18T17:00:00 2022-12-02T17:00:00

Online

Online

Optimal Policies for Dynamic Pricing and Inventory Control with Nonparametric Censored Demands

Boxiao (Beryl) Chen, University of Illinois, Chicago

We study the classic model of joint pricing and inventory control with lost sales over T consecutive review periods. The firm does not know the demand distribution a priori and needs to learn it from historical censored demand data. We develop nonparametric online learning algorithms that converge to the clairvoyant optimal policy at the fastest possible speed. The fundamental challenges rely on the fact that neither zeroth-order nor first-order feedbacks are accessible to the firm and reward at any single price is not observable due to demand censoring. We propose a novel inversion method based on empirical measures to consistently estimate the difference of the instantaneous reward functions at two prices, directly tackling the fundamental challenge brought by censored demands. Based on this technical innovation, we design bisection and trisection search methods that attain an $O(T^\{0.5\})$ regret for the case with concave reward functions, and we design an active tournament elimination method that attains $O(T^\{0.6\})$ regret when the reward functions are non-concave. We complement the $O(T^\{0.6\})$ regret upper bound with a matching regret lower bound, which is established by a novel information-theoretical argument based on generalized squared Hellinger distance.

Speaker Biography: 

Beryl Chen is an Associate Professor at the College of Business Administration, University of Illinois Chicago. In her research works, Beryl applies techniques from statistics, machine learning and optimization to develop data-driven algorithms for decision-making. Some of her works include dynamic pricing, inventory control and supply chain management, assortment planning, retailing, energy and capacity expansion.


If you have any further questions about the topic or speaker, please contact Jalaj Upadhyay at jalaj.upadhyay@rutgers.edu