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 approach without 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.

PDF, PS

In Leslie Pack Kaelbling, editors, Recent Advances in Reinforcement Learning(AI94-224), 11-32, 1996.

@Article{moriarty:mlj96, title={Efficient Reinforcement Learning Through Symbiotic Evolution}, author={David E. Moriarty and Risto Miikkulainen}, booktitle={Recent Advances in Reinforcement Learning}, journal={Machine Learning}, number={AI94-224}, editor={Leslie Pack Kaelbling}, institution={Department of Computer Sciences, The University of Texas at Austin}, pages={11-32}, url="http://nn.cs.utexas.edu/?moriarty:mlj96", year={1996} }