In this paper, a new synthesis approach is developed for associative memories based on the perceptron training algorithm. The design (synthesis) problem of feedback neural networks for associative memories is formulated as a set of linear inequalities such that the use of perceptron training is evident. The perceptron training in the synthesis algorithms is guaranteed to converge for the design of neural networks without any constraints on the connection matrix. For neural networks with constraints on the diagonal elements of the connection matrix, results concerning the properties of such networks and concerning the existence of such a network design are established. For neural networks with sparsity and/or symmetry constraints on the connection matrix, design algorithms are presented. Applications of the present synthesis approach to the design of associative memories realized by means of other feedback neural network models are studied. To demonstrate the applicability of the present results and to compare the present synthesis approach with existing design methods, specific examples are considered.

This paper presents a new synthesis procedure (design algorithm) for cellular neural networks (CNN's) with a space-invariant cloning template with applications to associative memories. In the present synthesis procedure, the design problem is formulated as a set of linear inequalities, and the inequalities are solved using the well-known perceptron training algorithm. When desired memory patterns are given by a set of bipolar vectors, it is guaranteed that a cellular neural network with a space-invariant cloning template can be designed using the design algorithm developed herein. An algorithm is also provided to design CNN's with space-invariant cloning templates and with symmetric connection matrices to guarantee the global stability of the network. Two specific examples are included to demonstrate the applicability of the methodology developed herein.Index Terms-Associative memories, cellular neural networks, perceptron training.

No abstract

No abstract

No abstract