Learning GMMs with Nearly Optimal Robustness Guarantees
Machine Learning
2021-11-16 v2 Data Structures and Algorithms
Statistics Theory
Machine Learning
Statistics Theory
Abstract
In this work we solve the problem of robustly learning a high-dimensional Gaussian mixture model with components from -corrupted samples up to accuracy in total variation distance for any constant and with mild assumptions on the mixture. This robustness guarantee is optimal up to polylogarithmic factors. The main challenge is that most earlier works rely on learning individual components in the mixture, but this is impossible in our setting, at least for the types of strong robustness guarantees we are aiming for. Instead we introduce a new framework which we call {\em strong observability} that gives us a route to circumvent this obstacle.
Cite
@article{arxiv.2104.09665,
title = {Learning GMMs with Nearly Optimal Robustness Guarantees},
author = {Allen Liu and Ankur Moitra},
journal= {arXiv preprint arXiv:2104.09665},
year = {2021}
}