Q-learning based on hierarchical evolutionary mechanism

Authors:
Masaya Yoshikawa;Takeshi Kihira;Hidekazu Terai
Affiliations:
Department of Information Engineering, Meijo University, Nagoya, Aichi, Japan;Department of VLSI System Design, Ritsumeikan University, Kusatsu, Shiga, Japan;Department of VLSI System Design, Ritsumeikan University, Kusatsu, Shiga, Japan
Venue:
WSEAS Transactions on Systems and Control
Year:
2008

Citing 9
Cited 2

Integrated architecture for learning, planning, and reacting based on approximating dynamic programming

Proceedings of the seventh international conference (1990) on Machine learning
Adaptation in natural and artificial systems

Adaptation in natural and artificial systems
Transfer of Learning by Composing Solutions of Elemental Sequential Tasks

Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning

Genetic Algorithms in Search, Optimization and Machine Learning
Using Transitional Proximity for Faster Reinforcement Learning

ML '92 Proceedings of the Ninth International Workshop on Machine Learning
Constraint-driven floorplanning based on genetic algorithm

CEA'07 Proceedings of the 2007 annual Conference on International Conference on Computer Engineering and Applications
Dedicated hardware for scheduling problems using genetic algorithm

AEE'06 Proceedings of the 5th WSEAS international conference on Applications of electrical engineering
Cooperative strategy based on adaptive Q-learning for robot soccer systems

IEEE Transactions on Fuzzy Systems
Combination of online clustering and Q-value based GA for reinforcement fuzzy system design

IEEE Transactions on Fuzzy Systems

Hardware accelerator for evolutionary robotics

WSEAS Transactions on Circuits and Systems
Grammar-based classifier system: a universal tool for grammatical inference

WSEAS Transactions on Computers

Quantified Score

Hi-index	0.00

Visualization

Abstract

Reinforcement learning is applied to various fields such as robotics and mechatronics control. The reinforcement learning is an efficient method to control in unknown environment. This paper discusses a new reinforcement learning algorithm which is based on Genetic Algorithm and has a hierarchical evolutionary mechanism. The proposed learning algorithm introduces new adaptive action value tables and it enables sharing knowledge among agents effectively. Regarding sharing knowledge among agents, the knowledge is inherited not only across the generations, but also in one generation. As a result, the proposed algorithm achieves effective learning, and realizes robustness learning. Computational simulations using the pong simulator which executes table tennis prove the effectiveness of the proposed algorithm.