Automatic generation of fuzzy inference systems by dynamic fuzzy Q-learning
Authors:
- 23 April 2004
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
This paper presents a dynamic Q-learning (DFQL) method that is capable of tuning the fuzzy inference systems (FIS) online. On-line self-organizing learning is developed so that structure and parameters identification are accomplished automatically and simultaneously based only on Q-learning. Self-organizing fuzzy inference is introduced to calculate actions and Q-functions so as to enable us to deal with continuous-valued states and actions. Fuzzy rules provide a natural mean to incorporate the bias components for rapid reinforcement learning. Experimental results and comparative studies with the fuzzy Q-learning the wall following task of mobile robots demonstrate the superiority of the proposed DFQL method.Keywords
This publication has 5 references indexed in Scilit:
- Continuous-Action Q-LearningMachine Learning, 2002
- A fast approach for automatic generation of fuzzy rules by generalized dynamic fuzzy neural networksIEEE Transactions on Fuzzy Systems, 2001
- Reinforcement Learning: An IntroductionIEEE Transactions on Neural Networks, 1998
- Fuzzy inference system learning by reinforcement methodsIEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 1998
- Incremental multi-step Q-learningMachine Learning, 1996