USE OF CMAC NEURAL NETWORKS IN REINFORCEMENT SELF-LEARNING CONTROL
暂无分享,去创建一个
This paper presents a technique that integrates the Cerebellar Model Articulation Controller (CMAC) into a self-learning control scheme developed by Barto et al. The modified technique has the learned information distributively stored. This prevents the waste of memory on useless states and reduces the memory size. The capability of information interpolation helps improve the learning speed.
[1] Richard S. Sutton,et al. Neuronlike adaptive elements that can solve difficult learning control problems , 1983, IEEE Transactions on Systems, Man, and Cybernetics.
[2] Donald E. Knuth,et al. Sorting and Searching , 1973 .
[3] James S. Albus,et al. New Approach to Manipulator Control: The Cerebellar Model Articulation Controller (CMAC)1 , 1975 .