An initialization mechanism is presented for Kohonen neural network implemented in CMOS technology. Proper selection of initial values of neurons' weights has a large influenc...
Max and min operations have interesting properties that facilitate the exchange of information between the symbolic and real-valued domains. As such, neural networks that employ m...
A new architecture and method for feature selection and extraction of logical rules from neural networks trained with backpropagation algorithm is presented. The network consists ...
Wlodzislaw Duch, Rafal Adamczak, Krzysztof Grabcze...
Dimension reduction for regression (DRR) deals with the problem of finding for high-dimensional data such low-dimensional representations, which preserve the ability to predict a ...
This paper presents an approach to the joint optimization of neural network structure and weights which can take advantage of backpropagation as a specialized decoder. The approach...