Intelligent Information Processing Laboratory
Building intelligent information engineering systems with the use of artificial intelligence
We are conducting research, for artificial intelligence, on a machine learning method called reinforcement learning and its applications to multi-agent systems. Reinforcement learning is a method used to gradually learn how to make optimal behavioral choices by giving rewards or punishments depending on the results of the learning so as to strengthen or relax the rules of behavioral selections. Multi-agent systems refer to a group of many individual agents interacting with each other.
| Affiliation | Computer Science and Engineering |
| Faculty Name | IGARASHI, Harukazu |
| Academic Society |
The Japanese Society for Artificial Intelligence The Institute of Electronics, Information and Communication Engineers Information Processing Society of Japan |
| Keyword | Artificial intelligence, Multi-agent, Computer shogi, Game development, Algorithms, Robo Cup |
Study Fields
- Information engineering
- Information science
For Society
Our research is eventually expected to be useful for managing systems that coordinate cooperative work using a group of robots and computers in areas such as making schedules (for freight transportation plans, etc.) and controlling systems (for traffics and unmanned transport vehicles, etc.) in the future.Research Themes
- Theoretical research on reinforcement learning
- Research on the control of agents in the robot soccer competition Robo Cup and machine learning of cooperative behaviors
- Checkerboard assessment and search control in shogi programs