English

Representational aspects of depth and conditioning in normalizing flows

Machine Learning 2021-06-29 v2 Machine Learning

Abstract

Normalizing flows are among the most popular paradigms in generative modeling, especially for images, primarily because we can efficiently evaluate the likelihood of a data point. This is desirable both for evaluating the fit of a model, and for ease of training, as maximizing the likelihood can be done by gradient descent. However, training normalizing flows comes with difficulties as well: models which produce good samples typically need to be extremely deep -- which comes with accompanying vanishing/exploding gradient problems. A very related problem is that they are often poorly conditioned: since they are parametrized as invertible maps from RdRd\mathbb{R}^d \to \mathbb{R}^d, and typical training data like images intuitively is lower-dimensional, the learned maps often have Jacobians that are close to being singular. In our paper, we tackle representational aspects around depth and conditioning of normalizing flows: both for general invertible architectures, and for a particular common architecture, affine couplings. We prove that Θ(1)\Theta(1) affine coupling layers suffice to exactly represent a permutation or 1×11 \times 1 convolution, as used in GLOW, showing that representationally the choice of partition is not a bottleneck for depth. We also show that shallow affine coupling networks are universal approximators in Wasserstein distance if ill-conditioning is allowed, and experimentally investigate related phenomena involving padding. Finally, we show a depth lower bound for general flow architectures with few neurons per layer and bounded Lipschitz constant.

Keywords

Cite

@article{arxiv.2010.01155,
  title  = {Representational aspects of depth and conditioning in normalizing flows},
  author = {Frederic Koehler and Viraj Mehta and Andrej Risteski},
  journal= {arXiv preprint arXiv:2010.01155},
  year   = {2021}
}

Comments

Appeared in ICML 2021

R2 v1 2026-06-23T18:59:01.498Z