Sciweavers

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

10

FOCS
1999
IEEE

favoriteEmaildiscussreport

102views Theoretical Computer Science» more FOCS 1999»

Learning Mixtures of Gaussians

13 years 8 months ago

Learning Mixtures of Gaussians

Download www.ml.inf.ethz.ch

Mixtures of Gaussians are among the most fundamental and widely used statistical models. Current techniques for learning such mixtures from data are local search heuristics with weak performance guarantees. We present the first provably correct algorithm for learning a mixture of Gaussians. This algorithm is very simple and returns the true centers of the Gaussians to within the precision specified by the user, with high probability. It runs in time only linear in the dimension of the data and polynomial in the number of Gaussians.

Sanjoy Dasgupta

Real-time Traffic

FOCS 1999 | Local Search Heuristics | Mixtures | Theoretical Computer Science | Weak Performance Guarantees |

claim paper

Related Content

» Manifold Constrained Finite Gaussian Mixtures

» GossipBased Greedy Gaussian Mixture Learning

» Bayesian Feature and Model Selection for Gaussian Mixture Models

» Boosted mixture learning of Gaussian mixture HMMs for speech recognition

» Simplifying Mixture Models Using the Unscented Transform

» A Greedy EM Algorithm for Gaussian Mixture Learning

» PAC Learning Mixtures of AxisAligned Gaussians with No Separation Assumption

» SAR images as mixtures of Gaussian mixtures

» Polynomial Learning of Distribution Families

Post Info
More Details (n/a)

Added	03 Aug 2010
Updated	03 Aug 2010
Type	Conference
Year	1999
Where	FOCS
Authors	Sanjoy Dasgupta

Comments (0)