On Learning Mixtures of Well-Separated Gaussians
2017 IEEE 58TH ANNUAL SYMPOSIUM ON FOUNDATIONS OF COMPUTER SCIENCE (FOCS), pp. 85.0-96, 2017.
mixtures of Gaussiansunsupervised learningclusteringparameter estimationsample complexityMore(1+)
We consider the problem of efficiently learning mixtures of a large number of spherical Gaussians, when the components of the mixture are well separated. In the most basic form of this problem, we are given samples from a uniform mixture of k standard spherical Gaussians with means mu(1),..., mu(k) is an element of R-d, and the goal is to...More
Full Text (Upload PDF)
PPT (Upload PPT)