Title :
NEFRL: A New Neuro-Fuzzy System for Episodic Reinforcement Learning Tasks
Author :
Behsaz, Babak ; Safabakhsh, Reza
Author_Institution :
Dept. of Comput. Eng. & Inf. Technol., Amirkabir Univ. of Technol., Tehran
Abstract :
In this paper, we propose a new neuro-fuzzy system for episodic reinforcement learning tasks, NEFRL. While NEFRL has all benefits of a neuro-fuzzy architecture, it has the additional advantage that it can learn with a numerical evaluation of performance and there is no need for training input-output pairs. Also, we show that the learning algorithm of this system converges with probability one to a local maximum of the average numerical performance signal. Our experimental results for the pole-balancing task show the power of this system even without any prior domain knowledge.
Keywords :
fuzzy neural nets; fuzzy systems; learning (artificial intelligence); NEFRL; episodic reinforcement learning tasks; learning algorithm; neurofuzzy system; pole-balancing task; probability; Computer architecture; Fellows; Fuzzy neural networks; Fuzzy systems; Gradient methods; Information technology; Machine learning; Neural networks; Stochastic processes; Unsupervised learning;
Conference_Titel :
Frontiers in the Convergence of Bioscience and Information Technologies, 2007. FBIT 2007
Conference_Location :
Jeju City
Print_ISBN :
978-0-7695-2999-8
DOI :
10.1109/FBIT.2007.139
