PhD Defense: Learning and Robustness With Applications to Mechanism Design

The design of economic mechanisms, especially auctions, is an increasingly important part of the modern economy. A particularly valuable property for a mechanism is strategyproofness: the mechanism must be robust to strategic manipulations so that the participants in the mechanism have no incentive to lie. Yet in the important case when the mechanism designer’s goal is to maximize their own revenue, the design of optimal strategyproof mechanisms has proved immensely difficult, with very little progress after decades of research. Recently, to escape this impasse, a number of works have parameterized auction mechanisms as deep neural networks and used gradient descent to successfully learn approximately optimal and approximately strategyproof mechanisms. We present several improvements on these techniques in order to ensure allocations satisfy desirable properties such as fairness, to ensure that violations of strategyproofness can be measured, and (with some compromises) to ensure that strategyproofness is always perfectly satisfied.
Examining Committee:

Chair:Co-Chair:Dean's Representative:Members:

Dr. John Dickerson Dr. Tom GoldsteinDr. Daniel Vincent Dr. Aravind Srinivasan Dr. Ian Kash (Univ. of IL-Chicago)