Random Walks with Tweedie: A Unified View of Score-Based Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Park, Chicago Y., McCann, Michael T., Garcia-Cardona, Cristina, Wohlberg, Brendt, Kamilov, Ulugbek S.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911189718532096
author Park, Chicago Y.
McCann, Michael T.
Garcia-Cardona, Cristina
Wohlberg, Brendt
Kamilov, Ulugbek S.
author_facet Park, Chicago Y.
McCann, Michael T.
Garcia-Cardona, Cristina
Wohlberg, Brendt
Kamilov, Ulugbek S.
contents We present a concise derivation for several influential score-based diffusion models that relies on only a few textbook results. Diffusion models have recently emerged as powerful tools for generating realistic, synthetic signals -- particularly natural images -- and often play a role in state-of-the-art algorithms for inverse problems in image processing. While these algorithms are often surprisingly simple, the theory behind them is not, and multiple complex theoretical justifications exist in the literature. Here, we provide a simple and largely self-contained theoretical justification for score-based diffusion models that is targeted towards the signal processing community. This approach leads to generic algorithmic templates for training and generating samples with diffusion models. We show that several influential diffusion models correspond to particular choices within these templates and demonstrate that alternative, more straightforward algorithmic choices can provide comparable results. This approach has the added benefit of enabling conditional sampling without any likelihood approximation.
format Preprint
id arxiv_https___arxiv_org_abs_2411_18702
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Random Walks with Tweedie: A Unified View of Score-Based Diffusion Models
Park, Chicago Y.
McCann, Michael T.
Garcia-Cardona, Cristina
Wohlberg, Brendt
Kamilov, Ulugbek S.
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Image and Video Processing
We present a concise derivation for several influential score-based diffusion models that relies on only a few textbook results. Diffusion models have recently emerged as powerful tools for generating realistic, synthetic signals -- particularly natural images -- and often play a role in state-of-the-art algorithms for inverse problems in image processing. While these algorithms are often surprisingly simple, the theory behind them is not, and multiple complex theoretical justifications exist in the literature. Here, we provide a simple and largely self-contained theoretical justification for score-based diffusion models that is targeted towards the signal processing community. This approach leads to generic algorithmic templates for training and generating samples with diffusion models. We show that several influential diffusion models correspond to particular choices within these templates and demonstrate that alternative, more straightforward algorithmic choices can provide comparable results. This approach has the added benefit of enabling conditional sampling without any likelihood approximation.
title Random Walks with Tweedie: A Unified View of Score-Based Diffusion Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Image and Video Processing
url https://arxiv.org/abs/2411.18702