David Pardoe


I completed my Ph.D. in spring 2011 in the Department of Computer Science at UT Austin. My advisor was Peter Stone.

Download my dissertation: PDF PS

Research Summary:

My research focuses on applications of machine learning in multi-agent systems, particularly in e-commerce settings. This research has been motivated by my participation in the Trading Agent Competition, where I have designed winning autonomous agents in both supply chain management and ad auction scenarios. My Ph.D. thesis explores methods by which agents in such settings can adapt to the behavior of other agents, with a particular focus on the use of transfer learning to learn quickly from limited interaction with these agents.

Information on our winning agents in the Trading Agent Competition in the domains of supply chain management and ad auctions can be found here.

Research interests by topic:

agent-based learning (supervised learning, regression, reinforcement learning, transfer learning), multi-agent learning, trading agents, auctions, e-commerce, game theory, mechanism design

Curriculum Vitae