LEARNING OPTIMAL CONJUNCTIVE CONCEPTS THROUGH A TEAM OF STOCHASTIC AUTOMATA

The problem of learning conjunctive concepts from a series of positive and negative examples of the concept is considered. Employing a proba bilistic structure on the domain, the goal of such inductive learning is precisely characterized. A parallel distributed stochastic algorith m is presented. It is proved that the algorithm will converge to the c oncept description with maximum probability of correct classification in the presence of up to 50% unbiased noise. A novel neural network st ructure that implements the learning algorithm is proposed. Through em pirical studies it is seen that the algorithm is quite efficient for l earning conjunctive concepts.