A feedforward network of learning automata for pattern classification

A model made of units of teams of learning automata is developed for the three layer pattern classifier. The algorithm is approximated by an ordinary differential equation (ODE), using weak convergence methods. The pattern recognition problem is posed as a constrained maximization problem. It is shown that the zeros of the ODE correspond to points satisfying first order necessary conditions of the maximization problem. Partial results are obtained showing that the ODE, and hence the algorithm, converge to the local maxima of the maximization problem