Energy-Based Models (EBMs) offer a versatile framework for modeling complex data distributions. However, training and sampling from EBMs continue to pose significant challenges. The widely-used Denoising Score Matching (DSM) method for scalable EBM training suffers from inconsistency issues, causing the energy model to learn a `noisy' data distribution. In this work, we propose an efficient sampling framework: (pseudo)-Gibbs sampling with moment matching, which enables effective sampling from the underlying clean model when given a `noisy' model that has been well-trained via DSM. We explore the benefits of our approach compared to related methods and demonstrate how to scale the method to high-dimensional datasets.
@article{arxiv.2305.11650,
title = {Moment Matching Denoising Gibbs Sampling},
author = {Mingtian Zhang and Alex Hawkins-Hooker and Brooks Paige and David Barber},
journal= {arXiv preprint arXiv:2305.11650},
year = {2024}
}