Cooperative updating in the Hopfield model

CONVERGENCE ANALYSIS OF A DISCRETE HOPFIELD NEURAL NETWORK WITH DELAY AND ITS APPLICATION TO KNOWLEDGE REFINEMENT

This paper investigates the convergence theorems that are associated with a Discrete Hopfield Neural Network (DHNN) with delay. We present two updating rules, one for serial mode and the other for parallel mode. The speed of convergence of these proposed updating rules is faster than all of the existing updating rules. It has been proved in this paper that a DHNN with delay will converge to a stable state when operating in a serial mode if the matrix of weights of the no-delay term is symmetric. In addition, it has been proved that they will converge to a stable state when operating in a parallel mode if the matrix of weights of the no-delay term is a symmetric and non-negative definite matrix. The condition for convergence of a DHNN without delay can been relaxed from the need to have a symmetric matrix to an even weaker condition of having a quasi-symmetric matrix. The results in this paper extend both the existing results concerning the convergence of a DHNN without delay and our previous findings. By means of the new network structure and its convergence theorems, we propose a local searching algorithm for combinatorial optimization. We also relate the maximum value of a bivariate energy function to the stable states of a DHNN with delay, which generalizes Hopfield's energy function. Moreover, for the serial model we give the relationship between the convergence of the energy function and the convergence of the corresponding network. One application is presented to demonstrate the higher rate of convergence and the accuracy of the classification using our algorithm.

The co-adaptive neural network approach to the Euclidean Travelling Salesman Problem

In this paper we consider the Euclidean Travelling Salesman Problem (ETSP). This is the problem of finding the shortest tour around a number of cities where the cities correspond to points in the Euclidean plane and the distances between cities are given by the usual Euclidean distance metric. We present a review of the literature with respect to neural network (NN) approaches for the ETSP, and the computational results that have been reported. Based upon this review we highlight two areas that are, in our judgement, currently neglected/lacking in the literature. These are: failure to make significant use of publicly available ETSP test problems in computational work, failure to address co-operation between neurons. Drawing upon our literature survey this paper presents a new Self-Organising NN approach, called the Co-Adaptive Net, which involves not just unsupervised learning to train neurons, but also allows neurons to co-operate and compete amongst themselves depending on their situation. Our Co-Adaptive Net algorithm also includes a number of algorithmic mechanisms that, based upon our literature review, we consider to have contributed to the computational success of previous algorithms. Results for 91 publicly available standard ETSP's are presented in this paper. The largest of these problems involves 85,900 cities. This paper presents: the most extensive computational evaluation of any NN approach on publicly available ETSP test problems that has been made to date in the literature, a NN approach that performs better, with respect to solution quality and/or computation time, than other NN approaches given previously in the literature. Drawing upon computational results produced as a result of the DIMACS TSP Challenge, we highlight the fact that none of the current NN approaches for the ETSP can compete with state of the art Operations Research heuristics. We discuss why we consider continuing to study and develop NN approaches for the ETSP to be of value.