Learning Mixtures of Gaussians in High Dimensions

TL;DR

This paper presents a polynomial-time algorithm for learning Gaussian mixture models in high dimensions under a smoothed analysis framework, overcoming worst-case hardness by leveraging tensor decomposition and structured random matrix analysis.

Contribution

It introduces a novel high-dimensional learning algorithm for Gaussian mixtures under smoothed analysis, utilizing tensor decomposition and new bounds on structured random matrices.

Findings

Learned Gaussian mixtures with polynomial complexity when n > Ω(k^2).

Developed new tensor decomposition techniques exploiting Gaussian moment symmetries.

Established bounds on smallest singular values of structured random matrices.

Abstract

Efficiently learning mixture of Gaussians is a fundamental problem in statistics and learning theory. Given samples coming from a random one out of k Gaussian distributions in Rn, the learning problem asks to estimate the means and the covariance matrices of these Gaussians. This learning problem arises in many areas ranging from the natural sciences to the social sciences, and has also found many machine learning applications. Unfortunately, learning mixture of Gaussians is an information theoretically hard problem: in order to learn the parameters up to a reasonable accuracy, the number of samples required is exponential in the number of Gaussian components in the worst case. In this work, we show that provided we are in high enough dimensions, the class of Gaussian mixtures is learnable in its most general form under a smoothed analysis framework, where the parameters are randomly perturbed from an adversarial starting point. In particular, given samples from a mixture of Gaussians with randomly perturbed parameters, when n > {\Omega}(k^2), we give an algorithm that learns the parameters with polynomial running time and using polynomial number of samples. The central algorithmic ideas consist of new ways to decompose the moment tensor of the Gaussian mixture by exploiting its structural properties. The symmetries of this tensor are derived from the combinatorial structure of higher order moments of Gaussian distributions (sometimes referred to as Isserlis' theorem or Wick's theorem). We also develop new tools for bounding smallest singular values of structured random matrices, which could be useful in other smoothed analysis settings.