Neural Net with Two Hidden Layers for Non-Linear Blind Source Separation

In this paper, we present an algorithm that minimizes the mutual information between the outputs of a perceptron with two hidden layers. The neural network is then used as separating system in the NonLinear Blind Source Separation problem.

• Type: Conference
• Year: 2003
• Where: ESANN
• Authors: Rubén Martín-Clemente, Susana Hornillo-Mellado, José I. Acha, Fernando Rojas Ruiz, Carlos García Puntonet
• Keywords: ESANN 2003, ESANN 2007, Hidden Layers, NonLinear Blind Source, Separation Problem