On Fuzzy e-Negotiation Agents: Autonomous negotiation with incomplete and imprecise information

Ryszard Kowalczyk
Applied Artificial Intelligence, CMIS

Friday, 28th April 2000
0930 - 1100

Conference Room
CSIRO Mathematical & Information Sciences
Ground Floor, 723 Swanston Street, Melbourne.

Abstract:

I will present some aspects of Fuzzy e-Negotiation Agents (FeNAs) for autonomous multi-issue negotiation in the presence of limited common knowledge and imprecise/soft constraints and preferences.  FeNAs use principles of utility theory and fuzzy constraint-based reasoning in order to find a consensus that maximizes an agent's utility at the highest possible level of fuzzy constraint satisfaction subject to its acceptability by other agents.  The capabilities of FeNAs for multi-issue integrative negotiation will be illustrated with two scenarios for negotiating document translation services and used car trading.  I'll also outline the applicability of the presented approach for multi-agent coordination and negotiation with varying level of common knowledge shared between the agents.

Biosketch:

Ryszard Kowalczyk is a Principal Research Scientist with CSIRO Mathematical and Information Sciences. His research interests include Soft Computing techniques (fuzzy systems, constraint-based reasoning and evolutionary computation) in collaborative and competitive decision making, with application to Agent Mediated e-Commerce.