Autonomous Bidding in the Trading Agent Competition

Amy Greenwald and Justin Boyan

Abstract

This paper asks the question: can adaptive, but not necessarily rational, agents learn Nash equilibrium behavior in the Santa Fe Bar Problem? To answer this question, three learning algorithms are simulated: fictitious play, no-regret learning, and $Q$-learning. Conditions under which these algorithms can converge to equilibrium behavior are isolated. But it is noted that the pure strategy Nash equilibria are unfair, while the (symmetric) mixed strategy equilibrium is inefficient. Thus, \sfbp\ is redesigned to induce adaptive agents to learn fair and efficient equilibrium outcomes.