Learning Mixtures of Gaussians

Sanjoy Dasgupta

EECS Department
University of California, Berkeley
Technical Report No. UCB/CSD-99-1047
May 1999

http://www.eecs.berkeley.edu/Pubs/TechRpts/1999/CSD-99-1047.pdf

We give the first provably correct algorithm for learning a mixture of Gaussians from data. The running time is only linear in the dimension of the data and polynomial in the number of Gaussians.


BibTeX citation:

@techreport{Dasgupta:CSD-99-1047,
    Author = {Dasgupta, Sanjoy},
    Title = {Learning Mixtures of Gaussians},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {1999},
    Month = {May},
    URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/1999/5673.html},
    Number = {UCB/CSD-99-1047},
    Abstract = {We give the first provably correct algorithm for learning a mixture of Gaussians from data. The running time is only linear in the dimension of the data and polynomial in the number of Gaussians.}
}

EndNote citation:

%0 Report
%A Dasgupta, Sanjoy
%T Learning Mixtures of Gaussians
%I EECS Department, University of California, Berkeley
%D 1999
%@ UCB/CSD-99-1047
%U http://www.eecs.berkeley.edu/Pubs/TechRpts/1999/5673.html
%F Dasgupta:CSD-99-1047