Sciweavers

NIPS
1994

A Non-linear Information Maximisation Algorithm that Performs Blind Separation

13 years 4 months ago
A Non-linear Information Maximisation Algorithm that Performs Blind Separation
A new learning algorithmis derived which performs online stochastic gradient ascent in the mutual informationbetween outputs and inputs of a network. In the absence of a priori knowledge about the `signal' and `noise' components of the input, propagation of information depends on calibrating network non-linearities to the detailed higher-order moments of the input density functions. By incidentally minimising mutual information between outputs, as well as maximising their individual entropies, the network `factorises' the input into independent components. As an example application, we have achieved near-perfect separation of ten digitally mixed speech signals. Our simulations lead us to believe that our network performs better at blind separation than the HeraultJutten network, re ecting the fact that it is derived rigorously from the mutual information objective.
Anthony J. Bell, Terrence J. Sejnowski
Added 02 Nov 2010
Updated 02 Nov 2010
Type Conference
Year 1994
Where NIPS
Authors Anthony J. Bell, Terrence J. Sejnowski
Comments (0)