Sciweavers

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

10

ICML
2006
IEEE

favoriteEmaildiscussreport

86views Machine Learning» more ICML 2006»

An investigation of computational and informational limits in Gaussian mixture clustering

14 years 5 months ago

An investigation of computational and informational limits in Gaussian mixture clustering

Download ttic.uchicago.edu

We investigate under what conditions clustering by learning a mixture of spherical Gaussians is (a) computationally tractable; and (b) statistically possible. We show that using principal component projection greatly aids in recovering the clustering using EM; present empirical evidence that even using such a projection, there is still a large gap between the number of samples needed to recover the clustering using EM, and the number of samples needed without computational restrictions; and characterize the regime in which such a gap exists.

Nathan Srebro, Gregory Shakhnarovich, Sam T. Rowei

Real-time Traffic

ICML 2006 | Machine Learning | Present Empirical Evidence | Principal Component Projection |

claim paper

Related Content

» Density Estimation Using Mixtures of Mixtures of Gaussians

» Computing Gaussian Mixture Models with EM Using Equivalence Constraints

» Robust estimation of gaussian mixtures from noisy input data

» Scalable Modelbased Clustering by Working on Data Summaries

» Probabilistic Models for Generating Modelling and Matching Image Categories

» Mixture Densities for Video Objects Recognition

» NonParametric Probabilistic Image Segmentation

» Regular Texture Analysis as Statistical Model Selection

» LatentDescriptor Clustering for Unsupervised POS Induction

Post Info
More Details (n/a)

Added	17 Nov 2009
Updated	17 Nov 2009
Type	Conference
Year	2006
Where	ICML
Authors	Nathan Srebro, Gregory Shakhnarovich, Sam T. Roweis

Comments (0)