How Does an Agent Learn to Negotiate?

Abstract

Negotiation has been extensively discussed in game-theoretic, economic, and management science literatures for decades. Recent growing interest in autonomous interacting software agents and their potential application in areas such as electronic commerce has given increased importance to automated negotiation. Evidence both from theoretical analysis and from observations of human interactions suggests that if decision makers can somehow take into consideration what other agents are thinking and furthermore learn during their interactions how other agents behave, their payoff might increase. In this paper, we propose a sequential decision making model of negotiation, called Bazaar. It provides an adaptive, multi-issue negotiation model capable of exhibiting a rich set of negotiation behaviors. Within the proposed negotiation framework, we model learning as a Bayesian belief update process. We prove that under certain conditions learning is indeed beneficial.