Autors: Gospodinova, E. G., Mihailova, C. M. Title: MATHEMATICAL MODELING FOR IMAGE RECOGNITION USING ARTIFICIAL INTELLIGENCE Keywords: mathematical modeling, neural network, image, algorithm Abstract: In this paper, we study a mathematical model for pattern recognition using neural networks. A feed-forward neural network was built with the capabilities of a graphics accelerator, based on the Levenberg-Marquardt method. The introduced model summarizes and proposes a new neural network training algorithm with Bayes regularization, Nguyen-Widrow initialization, and the early stopping and control method. This enables and greatly expands the efficiency of solving problems where pattern recognition is available. References Issue
|
Вид: пленарен доклад в национален форум с межд. уч., публикация в реферирано издание