Moment Matching Denoising Gibbs Sampling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Mingtian, Hawkins-Hooker, Alex, Paige, Brooks, Barber, David
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917617697030144
author Zhang, Mingtian
Hawkins-Hooker, Alex
Paige, Brooks
Barber, David
author_facet Zhang, Mingtian
Hawkins-Hooker, Alex
Paige, Brooks
Barber, David
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2305_11650
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Moment Matching Denoising Gibbs Sampling
Zhang, Mingtian
Hawkins-Hooker, Alex
Paige, Brooks
Barber, David
Machine Learning
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.
title Moment Matching Denoising Gibbs Sampling
topic Machine Learning
url https://arxiv.org/abs/2305.11650