Learning Polynomial Transformations
Abstract
We consider the problem of learning high dimensional polynomial transformations of Gaussians. Given samples of the form , where is hidden and is a function where every output coordinate is a low-degree polynomial, the goal is to learn the distribution over . This problem is natural in its own right, but is also an important special case of learning deep generative models, namely pushforwards of Gaussians under two-layer neural networks with polynomial activations. Understanding the learnability of such generative models is crucial to understanding why they perform so well in practice. Our first main result is a polynomial-time algorithm for learning quadratic transformations of Gaussians in a smoothed setting. Our second main result is a polynomial-time algorithm for learning constant-degree polynomial transformations of Gaussian in a smoothed setting, when the rank of the associated tensors is small. In fact our results extend to any rotation-invariant input distribution, not just Gaussian. These are the first end-to-end guarantees for learning a pushforward under a neural network with more than one layer. Along the way, we also give the first polynomial-time algorithms with provable guarantees for tensor ring decomposition, a popular generalization of tensor decomposition that is used in practice to implicitly store large tensors.
Cite
@article{arxiv.2204.04209,
title = {Learning Polynomial Transformations},
author = {Sitan Chen and Jerry Li and Yuanzhi Li and Anru R. Zhang},
journal= {arXiv preprint arXiv:2204.04209},
year = {2022}
}
Comments
121 pages, comments welcome