Nonlinear Source Separation

Nonlinear source separation is the subject of much recent study, although until now no complete overview has been available. Author Luis Almeida fills this gap by providing a comprehensive look at the state of the art for practicing scientists and graduate students. Foundations and a detailed overview for nonlinear source separation methods are provided, with thorough references for readers who need more information on implementation of a specific method.Readers are assumed to have a basic knowledge of nonlinear trainable systems, including multilayer perceptrons (MLPs), and also the backpropagationtraining method and gradient-based optimization.Key Features:* A detailed introduction to nonlinear source separation* A discussion of the post-nonlinear setting, with theoretical results guaranteeing source recovery* A review of unconstrained separation methods, including MISEP, ensemble learning, and kTDSEPAlmeida, Luis B. is the author of 'Nonlinear Source Separation ', published 2004 under ISBN 9781598290301 and ISBN 1598290304.