10
64

Robustly Learning Mixtures of kk Arbitrary Gaussians

Abstract

We give a polynomial-time algorithm for the problem of robustly estimating a mixture of kk arbitrary Gaussians in Rd\mathbb{R}^d, for any fixed kk, in the presence of a constant fraction of arbitrary corruptions. This resolves the main open problem in several previous works on algorithmic robust statistics, which addressed the special cases of robustly estimating (a) a single Gaussian, (b) a mixture of TV-distance separated Gaussians, and (c) a uniform mixture of two Gaussians. Our main tools are an efficient \emph{partial clustering} algorithm that relies on the sum-of-squares method, and a novel \emph{tensor decomposition} algorithm that allows errors in both Frobenius norm and low-rank terms.

View on arXiv
Comments on this paper