Results 1 
3 of
3
Adaptive blind signal processingneural network approaches
 Proc. of the IEEE
, 1998
"... Learning algorithms and underlying basic mathematical ideas are presented for the problem of adaptive blind signal processing, especially instantaneous blind separation and multichannel blind deconvolution/equalization of independent source signals. We discuss recent developments of adaptive learnin ..."
Abstract

Cited by 61 (9 self)
 Add to MetaCart
(Show Context)
Learning algorithms and underlying basic mathematical ideas are presented for the problem of adaptive blind signal processing, especially instantaneous blind separation and multichannel blind deconvolution/equalization of independent source signals. We discuss recent developments of adaptive learning algorithms based on the natural gradient approach and their properties concerning convergence, stability, and efficiency. Several promising schemas are proposed and reviewed in the paper. Emphasis is given to neural networks or adaptive filtering models and associated online adaptive nonlinear learning algorithms. Computer simulations illustrate the performances of the developed algorithms. Some results presented in this paper are new and are being published for the first time.
Neural Codes and Independent Component Analysis: Information Theoretic Approach and Conditions on Cumulants
"... In this contribution we review recent results obtained on blind source separation (BSS) and independent component analysis (ICA). In particular we show that maximisation of mutual information can lead to ICA, and we present new conditions on cross cumulants which guarantee that blind source separati ..."
Abstract
 Add to MetaCart
In this contribution we review recent results obtained on blind source separation (BSS) and independent component analysis (ICA). In particular we show that maximisation of mutual information can lead to ICA, and we present new conditions on cross cumulants which guarantee that blind source separation has been performed. TAINN'97, Ankara, may 1997 1 Introduction Independent Component Analysis (ICA), and in particular Blind Source Separation (BSS), can be obtained from the maximization of mutual information, as first shown in [1]. This result was obtained for a deterministic processing system, with an arbitrary inputoutput relationship. The relevance for BSS was stressed out: in the particular case where the inputs are linear combinations of independent random variables ("sources"), one can use a feedforward network (with no hidden layer), and nonlinear transfer functions; then the outputs of the system will give the independent components if both the weigths and the transfer functio...
nonGaussian sources
, 1998
"... An algebraic principle for blind separation of white ..."
(Show Context)