conftrace_

2007 JMLR JMLR 2007

A Probabilistic Analysis of EM for Mixtures of Separated, Spherical Gaussians

Abstract

We show that, given data from a mixture of k well-separated spherical Gaussians in ℜd, a simple two-round variant of EM will, with high probability, learn the parameters of the Gaussians to near-optimal precision, if the dimension is high (d >> ln k). We relate this to previous theoretical and empirical work on the EM algorithm. [abs] [ pdf ][ bib ] © JMLR 2007. (edit, beta)

Authors

Sanjoy Dasgupta, Leonard Schulman

Topics

Machine Learning > Core Methods > Clustering Machine Learning > Optimization & Theory > Bayesian Inference Machine Learning > Optimization & Theory > Statistical Learning Machine Learning > Bayesian & Probabilistic > Probabilistic Modeling

Keywords

parameter estimation expectation maximization high-dimensional learning mixture model gaussian mixture probabilistic analysis spherical gaussian

More from JMLR 2007

All JMLR 2007 papers →

Minimax Regret Classifier for Imprecise Class Distributions

Distances between Data Sets Based on Summary Statistics

Building Blocks for Variational Bayesian Learning of Latent Variable Models

Noise Tolerant Variants of the Perceptron Algorithm

Learnability of Gaussians with Flexible Variances

Separating Models of Learning from Correlated and Uncorrelated Data