Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Joint Approximation Diagonalization of Eigen-matrices
open-in-new
<p>Joint Approximation Diagonalization of Eigen-matrices (JADE) is an algorithm for <a href="/facts/Independent_component_analysis/gsdVaDFR">independent component analysis</a> that separates observed mixed signals into <a href="/facts/Latent_variable/ohg99BnW">latent</a> source signals by exploiting <a href="/facts/Moment_(mathematics)/GL7vJ1nb">fourth order moments</a>. The fourth order moments are a measure of non-Gaussianity, which is used as a proxy for <a href="/facts/Independent_component_analysis/gsdVaDFR">defining independence between the source signals</a>. The motivation for this measure is that Gaussian distributions possess zero excess <a href="/facts/Kurtosis/yqGgry89">kurtosis</a>, and with non-Gaussianity being a canonical assumption of ICA, JADE seeks an orthogonal rotation of the observed mixed vectors to estimate source vectors which possess high values of excess kurtosis. </p>