Efficient Reinforcement Learning through Symbiotic Evolution

DMCA

Efficient Reinforcement Learning through Symbiotic Evolution (1996)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by David E. Moriarty , Risto Miikkulainen , Pack Kaelbling

Venue:	Machine Learning
Citations:	161 - 38 self

BibTeX

@INPROCEEDINGS{Moriarty96efficientreinforcement,
    author = {David E. Moriarty and Risto Miikkulainen and Pack Kaelbling},
    title = {Efficient Reinforcement Learning through Symbiotic Evolution},
    booktitle = {Machine Learning},
    year = {1996},
    pages = {11--32}
}

OpenURL

Abstract

. This article presents a new reinforcement learning method called SANE (Symbiotic, Adaptive Neuro-Evolution), which evolves a population of neurons through genetic algorithms to form a neural network capable of performing a task. Symbiotic evolution promotes both cooperation and specialization, which results in a fast, efficient genetic search and discourages convergence to suboptimal solutions. In the inverted pendulum problem, SANE formed effective networks 9 to 16 times faster than the Adaptive Heuristic Critic and 2 times faster than Q- learning and the GENITOR neuro-evolution approachwithout loss of generalization. Such efficient learning, combined with few domain assumptions, make SANE a promising approach to a broad range of reinforcement learning problems, including many real-world applications. Keywords: Neuro-Evolution, Reinforcement Learning, Genetic Algorithms, Neural Networks. 1. Introduction Learning effective decision policies is a difficult problem that appears in m...

Keyphrases

symbiotic evolution efficient reinforcement learning genetic algorithm neural network effective network reinforcement learning problem effective decision policy reinforcement learning many real-world application promising approach new reinforcement learning method adaptive heuristic critic efficient genetic search inverted pendulum problem domain assumption difficult problem efficient learning broad range suboptimal solution adaptive neuro-evolution genitor neuro-evolution approachwithout loss

DMCA