De. Thompson et S. Kwon, NEIGHBORHOOD SEQUENTIAL AND RANDOM TRAINING TECHNIQUES FOR CMAC, IEEE transactions on neural networks, 6(1), 1995, pp. 196-202
An adaptive control algorithm based on Albus' CMAC (Cerebellar Model A
rticulation Controller) was studied with emphasis on how to train CMAC
systems. Two training techniques-neighborhood sequential training and
random training, have been devised, These techniques were used to gen
erate mathematical functions, and both methods successfully circumvent
ed the training interference resulting from CMAC's inherent generaliza
tion property, In the neighborhood sequential training method, a strat
egy was devised to utilize the discrete, finite state nature of the CM
AC's address space for selecting points in the input space which would
train CMAC systems in the most rapid manner possible. The random trai
ning method was found to converge on the training function with the gr
eatest precision, although it requires longer training periods than th
e neighborhood sequential training method to achieve a desired perform
ance level.