[CS] Minbiao Han Candidacy Exam/May 17, 2024

[meganwoodward at uchicago.edu]

This is an announcement of Minbiao Han's Candidacy Exam.
===============================================
Candidate: Minbiao Han

Date: Friday, May 17, 2024

Time:  1 pm CT

Remote Location: https://uchicago.zoom.us/j/97406340075?pwd=aCs4djF6RHg1UllhdHdHK1dJM3VDZz09 Meeting ID: 974 0634 0075. Passcode: 739782

Location: JCL 257

Title: Towards An Algorithmic Theory of Principal-Agency

Abstract: Principal-agent problems are fundamental to understanding many strategic interactions in practice. The economics literature has provided a rich theoretical foundation for principal-agent models, and the recent increase in attention from the computer science community has introduced fresh insights and new challenges. In addition, as we build upon these foundations laid by traditional approaches, we recognize that some may benefit from further refinement to capture the subtleties of agent decision-making, including bounded rationality and strategic deception, which can have a crucial impact on principal-agent relationships. By integrating perspectives from both economics and computer science, we aim to bridge the existing gaps and uncover a more comprehensive understanding of principal-agent problems with their real-world implications. 

In the first part of the proposed work, we study a ubiquitous learning challenge in online principal-agent problems, where the principal learns the agent's private information from their revealed preferences in past interactions. This paradigm includes important special cases such as pricing and contract design, which have been widely studied in recent literature. Secondly, to account for the subtleties of agent decision-making, we move beyond the myopic agent behavior model to accommodate strategic and dynamic interactions. Finally, we develop computational results on computing optimal and robust principal policies, which we then apply to a diverse range of real-world applications, including contract theory, market design, security games, and federated learning.

Advisors: Haifeng Xu

Committee Members: Haifeng Xu, Yuxin Chen, Cong Ma, and Michael Albert