Dynamic Assortment Planning under Discrete Choice Models.
|
Speaker (s):
WANG Yining
PhD Candidate,
Carnegie Mellon University
|
|
Date:
Time:
Venue:
|
|
December 20, 2018, Thursday
10:00am - 11:00am
Meeting Room 5.1, Level 5
School of Information Systems
Singapore Management University
80 Stamford Road
Singapore 178902
|
|
ABSTRACT
In this talk, I will present my works (joint with Xi Chen and Yuan Zhou) on dynamic assortment planning with various discrete choice models, including the plain multinomial logit (MNL) model and the linear contextual MNL model. For each arriving customer, the seller offers an assortment of substitutable products, and then the customer makes purchases according to pre-specified choice models. Since all utility parameters of customers are unknown, the seller needs to simultaneously learn customers' choice behavior and make dynamic decisions on assortments based on the current knowledge. For both the plain MNL and the contextual MNL discrete choice models, we develop efficient dynamic policy and establish near-optimal upper bounds on their corresponding regret, an important evaluation measure of the expected revenue of the planned assortments. Numerical results of our proposed policies will also be presented.
About the Speaker
Yining Wang is currently a 5th-year PhD student in the Machine Learning Department of Carnegie Mellon University. His major research interests are active learning, online learning and bandit optimization methods, as well as their applications to revenue management problems such as assortment optimization and dynamic pricing.
