Learning in multi-agent systems
In the recent years, multi-agent systems have gained increasing attention. Such systems can be cooperative or competitive. When designing multi-agent systems, designers are generally not able to tell an agent what to do in advance since it is impossible to predict all the situations an agent may exp...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Theses and Dissertations |
Published: |
2008
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/3759 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Summary: | In the recent years, multi-agent systems have gained increasing attention. Such systems can be cooperative or competitive. When designing multi-agent systems, designers are generally not able to tell an agent what to do in advance since it is impossible to predict all the situations an agent may experience. Agents have to learn. Although there are standard learning techniques, they need to be customized for a specific application domain. The main objective of this research is to investigate learning techniques available in a multi-agent setting, understand the cooperative and competitive learning, and develop and apply a new learning technique in a multi- agent area. |
---|