Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Kernel-independent component analysis
open-in-new
<p>In statistics, kernel-independent component analysis (kernel ICA) is an efficient algorithm for <a href="/facts/Independent_component_analysis/gsdVaDFR">independent component analysis</a> which estimates source components by optimizing a <i>generalized variance</i> contrast function, which is based on representations in a <a href="/facts/Reproducing_kernel_Hilbert_space/VegkYAXb">reproducing kernel Hilbert space</a>. Those contrast functions use the notion of mutual information as a <a href="/facts/Independent_component_analysis/gsdVaDFR">measure</a> of <a href="/facts/Independence_(probability_theory)/NUzQtnUL">statistical independence</a>. </p>