English

Verlet Flows: Exact-Likelihood Integrators for Flow-Based Generative Models

Machine Learning 2024-05-07 v1

Abstract

Approximations in computing model likelihoods with continuous normalizing flows (CNFs) hinder the use of these models for importance sampling of Boltzmann distributions, where exact likelihoods are required. In this work, we present Verlet flows, a class of CNFs on an augmented state-space inspired by symplectic integrators from Hamiltonian dynamics. When used with carefully constructed Taylor-Verlet integrators, Verlet flows provide exact-likelihood generative models which generalize coupled flow architectures from a non-continuous setting while imposing minimal expressivity constraints. On experiments over toy densities, we demonstrate that the variance of the commonly used Hutchinson trace estimator is unsuitable for importance sampling, whereas Verlet flows perform comparably to full autograd trace computations while being significantly faster.

Cite

@article{arxiv.2405.02805,
  title  = {Verlet Flows: Exact-Likelihood Integrators for Flow-Based Generative Models},
  author = {Ezra Erives and Bowen Jing and Tommi Jaakkola},
  journal= {arXiv preprint arXiv:2405.02805},
  year   = {2024}
}

Comments

ICLR AI4DifferentialEqautions In Science workshop 2024

R2 v1 2026-06-28T16:16:56.074Z