In this paper the strategy of Hardheaded negotiating agent is described. Our agent won the Automated Negotiating Agents Competition 2011. As the name implies, the agent is hardheaded, it will not concede until the very end. Using a concession function, it generates bids in a monotonic way, which resets to a random value after the dynamic concession limit is reached. In practice, this means that most of the time the agent will cycle through the same range of bids. Since the preferences of the opponent are not known, the agent tries to learn the opponent’s preference profile. It chooses bids which it thinks are optimal for the opponent in case there are equivalent bids for itself.
KeywordsDiscount Factor Learning Module Learning Function Minimum Utility Threshold Negotiate Agent
Unable to display preview. Download preview PDF.
- 1.Fatima, S.S., Wooldridge, M., Jennings, N.R.: Optimal Negotiation Strategies for Agents with Incomplete Information. In: Meyer, J.-J.C., Tambe, M. (eds.) ATAL 2001. LNCS (LNAI), vol. 2333, pp. 377–392. Springer, Heidelberg (2002)Google Scholar