Polynomial Discrete Time Cellular Neural Networks to solve the XOR problem
Eduardo Gomez-Ramirez, Giovanni Egidio Pazienza, X. Vilasis-Cardona · 2006
Some papers discuss different options to improve the capabilities of cellular neural networks (CNN). The principal point is that a single layer CNN can not solve problems with linearly nonseparable data. In this paper a new model called polynomial discrete time cellular neural networks is presented. This model has a very simple nonlinear term that can improve the performance of the network. The results show how it is possible to solve the XOR problem. The templates of the entire network are computed using genetic algorithm