----------------------------------------------------------------------------------------------------

    Department of Systems Engineering and Engineering Management

                                                        &

                    Department of Decisions, Operations and Technology

----------------------------------------------------------------------------------------------------



Date: Friday, September 13, 2024, 4:30 pm to 5:30 pm HKT

Venue: ERB513, The Chinese University of Hong Kong

Title: Semi-Separable Mechanisms in Multi-Item Robust Screening

Speaker: Prof. Shixin Wang, Department of Decisions, Operations and

Technology at The Chinese University of Hong Kong (CUHK) Business School



Abstract:
Consumers often find themselves overwhelmed by extensive assortments offered by online retailers and show bounded rationality behavior. However existing literature on dynamic assortment optimization didn’t consider consumers' such bounded rationality behavior. This motivates us to employ a simple and effective two-stage consider-then-choose model namely the Threshold Multinomial Logit (TMNL) model to investigate the online assortment optimization problem. The TMNL model characterizes consumers' endogenous consideration sets formation by the threshold effect. This endogenous dependency can capture more flexible substitution patterns than the classical MNL choice model but it also creates great difficulties for online learning. In the offline assortment setting we analyze the properties of optimal assortment and propose an efficient assortment optimization algorithm outperforms the benchmark. In the online setting with unknown customer preferences and consideration set formation we propose online learning algorithms that achieve nearly optimal regret bounds under both instance-independent and instance-dependent conditions. To the best of our knowledge we are the first work to consider online assortment problem with consumers' endogenous consider-then-choose behavior. Moreover our algorithm is extended to the contextual learning setting effectively mitigating the impact of the number of products on its performance. Extensive numerical experiments validate the efficacy of our proposed algorithms. This is joint works with Wenxiang Chen, Caihua Chen, Ruxian Wang and Weili Xue.

Biography:
Dr. Shen received B.S.(1989) in Mathematics from Nanjing University. He received a Ph.D. from Southeast University in 1995. He is currently  a professor in the School of Management & Engineering and the Department of Mathematics at Nanjing University. His research areas include data-driven decision-making, revenue management, supply chain management.

Everyone is welcome to attend the talk!

SEEM-5201 Website: https://seminar.se.cuhk.edu.hk

Email: seem5201@se.cuhk.edu.hk