MACC'97 Poster Session
- Learning Message Control in Multiagent Systems
- Toshiharu Sugawara, Satoshi Kurihara
- NTT
- Contact to: sugawara@brl.ntt.co.jp
- Abstract
This paper introduces the learning mechanism by which agents can
identify through experience important messages in the context of inference
at a specific situation. At first, agents may not be able to immediately
read and process the important messages because of inappropriate ratings
and incompleteness of knowledge for coordinated actions. By analyzing the
history of the past inferences with other agents, however, they identify
which messages were really used. Agents then generate situation-specific
rules for understanding important messages. Finally, an example for
explaining how agents can generate the control rule is also described.
- keywords
Multiagent Learning, Reasoning about coordinated interactions
Return to top page
Wed Jan 21 09:37:36 JST 1998