Gaussian mixture models and the EM algorithm

← Back to document page

Gaussian mixture models and the EM algorithmRamesh Sridharan These notes give a short introduction to Gaussian mixture models (GMMs) and theExpectation-Maximization (EM) algorithm , first for the specific case of GMMs, and thenmore generally. These notes assume you re familiar with basic probability and basic you re interested in the full derivation (Section 3), some familiarity with entropy and KLdivergence is useful but not strictly notation here is borrowed fromIntroduction to Probabilityby Bertsekas & Tsitsiklis:random variables are represented with capital letters, values they take are represented withlowercase letters,pXrepresents a probability distribution for random variableX, andpX(x)represents the probability of valuex(according topX). We ll also use the shorthand notationXn1to represent the sequenceX1,X2.

Gaussian mixture models and the EM algorithm Ramesh Sridharan These notes give a short introduction to Gaussian mixture models (GMMs) and the Expectation-Maximization (EM) algorithm, rst for the speci c case of GMMs, and then more generally. These notes assume you’re familiar with basic probability and basic calculus.

  Model