A structured approach to the design of neural-like learning automata. A novel fuzzy neuron is used as a basic processing element in building such automata. The possibility of implementing neural networks by VLSI circuits is explored. It is shown that a parallel automaton can be realized directly from specified fuzzy production rules. By reinforcing the fuzzy grades of membership of production rules, a fuzzy learning machine is obtained. This approach can also be applied to the construction of stochastic neural automata by using multiply-add operations instead of max-min rules.
|Original language||English (US)|
|Title of host publication||Unknown Host Publication Title|
|Editors||Maureen Caudill, Charles T. Butler, San Diego Adaptics|
|Place of Publication||San Diego, CA, USA|
|State||Published - 1987|
ASJC Scopus subject areas