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

             Department of Systems Engineering and Engineering Management
                     The Chinese University of Hong Kong

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

Date:Friday, April 8, 2022, 4:30 PM HKT

Title:Risk Preference and Time Preference Elicitations in Decision-Making

Speaker: Professor Wenjie Huang, The University of Hong Kong

Abstract:

Preference robust optimization (PRO) is a recent direction of robust optimization where ambiguity arises due to the decision maker (DM)’s inability to precisely articulate its own preferences, a lack of accurate models for human behavior, or scarce observations of the DM’s past choices.

In this work, we first introduce several existing methods on risk preference elicitation. Second, we use case studies to clarify the importance of an accurate time preference (i.e., discount factor) in sequential decision-making problems and introduce two existing field experiments with their observed DM’s behaviors. A principled constructive model is proposed that explains the behaviors and brings new managerial insights. The model prescribes actions that account for the subjective stopping time, and can be further reformulated as a recursive compounding model (C-model) on the probabilities of stopping. We derive relevant theoretical results between C-model with existing discounted dynamic risk models. Finally, field experiments and illustrative example are conducted to support C-model and show its impacts on decision-making.

This talk also raises two open questions: (1) how C-model can be applied in Markov decision processes, reinforcement learning (RL) and inverse RL; (2) in general, what is the link between PRO and distributionally robust optimization.

Biography:

Wenjie Huang is Research Assistant Professor in Department of Industrial and Manufacturing Systems Engineering, and Musketeers Foundation Institute of Data Science, The University of Hong Kong (HKU). Before joining HKU, he held joint postdoc positions at School of Data Science, The Chinese University of Hong Kong, Shenzhen and Group for Research in Decision Analysis (GERAD), Canada (from Sep 2019 to Sep 2021). He received Ph.D. degree from the Department of Industrial Systems Engineering and Management, National University of Singapore in June 2019.

His research interests are quantitative methodologies in decision-making under uncertainty, data-driven decision-making and sequential decision-making, with applications in smart society and operations management.