Abstract
A new neural network of adaptive resonance theory, which embodies the principles of associative memory and can be retrieved by associative input information to each other and the input image data, is designed. New neural network which allows to solve problems with multiple solutions, has the properties of additional training, stable and compact storage previously stored information, and this allows to use the network for the development of specialized knowledge bases which use associative information.
Keywords
Neural network of adaptive resonance theory, associative memory, problems with multiple solutions, additional training.
Dmitrienko, V.D. and Zakovorotnyi, A.Yu., 2014. Neural association memory of adaptive resonance theory that allows to define multiple solutions. Software of systems in the industrial and social fields, 1: 8-17.