Learning in Deep Factor Graphs with Gaussian Belief Propagation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Nabarro, Seth, van der Wilk, Mark, Davison, Andrew J
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929423148646400
author Nabarro, Seth
van der Wilk, Mark
Davison, Andrew J
author_facet Nabarro, Seth
van der Wilk, Mark
Davison, Andrew J
contents We propose an approach to do learning in Gaussian factor graphs. We treat all relevant quantities (inputs, outputs, parameters, latents) as random variables in a graphical model, and view both training and prediction as inference problems with different observed nodes. Our experiments show that these problems can be efficiently solved with belief propagation (BP), whose updates are inherently local, presenting exciting opportunities for distributed and asynchronous training. Our approach can be scaled to deep networks and provides a natural means to do continual learning: use the BP-estimated parameter marginals of the current task as parameter priors for the next. On a video denoising task we demonstrate the benefit of learnable parameters over a classical factor graph approach and we show encouraging performance of deep factor graphs for continual image classification.
format Preprint
id arxiv_https___arxiv_org_abs_2311_14649
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Learning in Deep Factor Graphs with Gaussian Belief Propagation
Nabarro, Seth
van der Wilk, Mark
Davison, Andrew J
Machine Learning
We propose an approach to do learning in Gaussian factor graphs. We treat all relevant quantities (inputs, outputs, parameters, latents) as random variables in a graphical model, and view both training and prediction as inference problems with different observed nodes. Our experiments show that these problems can be efficiently solved with belief propagation (BP), whose updates are inherently local, presenting exciting opportunities for distributed and asynchronous training. Our approach can be scaled to deep networks and provides a natural means to do continual learning: use the BP-estimated parameter marginals of the current task as parameter priors for the next. On a video denoising task we demonstrate the benefit of learnable parameters over a classical factor graph approach and we show encouraging performance of deep factor graphs for continual image classification.
title Learning in Deep Factor Graphs with Gaussian Belief Propagation
topic Machine Learning
url https://arxiv.org/abs/2311.14649