Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

10

JMLR
2007

favoriteEmaildiscussreport

87views more JMLR 2007»

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

13 years 4 months ago

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

Download jmlr.csail.mit.edu

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 nearoptimal precision, if the dimension is high (d lnk). We relate this to previous theoretical and empirical work on the EM algorithm.

Sanjoy Dasgupta, Leonard J. Schulman

Real-time Traffic

Empirical Work | JMLR 2007 | Simple Two-round Variant | Well-separated Spherical Gaussians |

claim paper

Related Content

» EM algorithms of Gaussian mixture model and hidden Markov model

» Integrating binaural cues and blind source separation method for separating reverberant sp...

» Toward Learning Gaussian Mixtures with Arbitrary Separation

» Enhanced Text Extraction from Arabic Degraded Document Images Using EM Algorithm

» Speech Enhancement Using Gaussian Scale Mixture Models

» Probabilistic SpatialTemporal Segmentation of Multiple Sclerosis Lesions

» Generative modelbased clustering of directional data

» Realcoded crossover as a role of kernel density estimation

» A Statistical Approach for Intensity Loss Compensation of Confocal Microscopy Images

Post Info
More Details (n/a)

Added	15 Dec 2010
Updated	15 Dec 2010
Type	Journal
Year	2007
Where	JMLR
Authors	Sanjoy Dasgupta, Leonard J. Schulman

Comments (0)