Intelligent Agents for Multi-Issue Auctions and Bidding

J. Debenham and S. Simoff (Australia)


Intelligent Agents. E-commerce.


An approach to auctions and bidding is founded on obser vations and expectations of the opponents’ behavior and not on assumptions concerning the opponents’ motivations or internal reasoning. The approach draws ideas from in formation theory. A bidding agent employs maximum en tropy inference to determine its actions on the basis of this uncertain data. Maximum entropy inference may be ap plied both to multi-issue and to single-issue negotiation. Multi-issue variants of the four common auction mecha nisms are discussed.

Important Links:

Go Back