Detail of Search result

Information Industrial Revolution in The Creative Technology
Intelligent Information Processing Laboratory
Harukazu Igarashi
Professor, Department of Information Science and Engineering
College of Engineering
Electrical Engineering and Computer Science in Master's Program
Functional Control Systems in Doctoral Program
Location: Toyosu Campus
Laboratory Overview
Reinforcement Learning and Multiagent system
We do research on reinforcement learning and its application to multiagent systems. Reinforcement learning make agent's policies by intensifying and supressing action rules through rewards and penalties given to agent's try and error processes. A mutiagent system is a set of agents that interact with each other and applied to designing intelligent systems.
Laboratory Character Mutiagent systems are drawing attention of researchers to design intelligent contorolling systems in many research fields. Reinforcement learning is very appropriate to make agents learn or adapt to new environments that change dynamically. It is used to make intelligent game programs in Chess, Go and Shogi.
Research Outputs You can review research papers in the website below.
Laboratory Research Field
RoboCup Computer Shogi Reinforcement Learning Multiagent System  
Recent Research Topics
Multiagent learning Monte Carlo softmax search Velocity control of a car based on Fuzzy control and reinforcement learning    
Affiliated Conferences
The Japanese Society for Artificial Intelligence Information Processing Society of Japan The Institute of Electronics Information and Communication Engineers  
For Social Contributions to the World Sustainability In future years, our research will contribute to system management of multiple robots and computers collaborating with each other. An example of its application is a scheduling system that controls automatic guided vehicles safely and effectively.
Related Industries
Applicable Industrial Fields
Instruments and Devices in this Laboratory
Research Seeds