Geodesic Variational Bayes for Multiway Covariances

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Simonis, Quinn, Wells, Martin T.
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916558390951936
author Simonis, Quinn
Wells, Martin T.
author_facet Simonis, Quinn
Wells, Martin T.
contents This article explores the optimization of variational approximations for posterior covariances of Gaussian multiway arrays. To achieve this, we establish a natural differential geometric optimization framework on the space using the pullback of the affine-invariant metric. In the case of a truly separable covariance, we demonstrate a joint approximation in the multiway space outperforms a mean-field approximation in optimization efficiency and provides a superior approximation to an unstructured Inverse-Wishart posterior under the average Mahalanobis distance of the data while maintaining a multiway interpretation. We moreover establish efficient expressions for the Euclidean and Riemannian gradients in both cases of the joint and mean-field approximation. We end with an analysis of commodity trade data.
format Preprint
id arxiv_https___arxiv_org_abs_2501_04935
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Geodesic Variational Bayes for Multiway Covariances
Simonis, Quinn
Wells, Martin T.
Computation
This article explores the optimization of variational approximations for posterior covariances of Gaussian multiway arrays. To achieve this, we establish a natural differential geometric optimization framework on the space using the pullback of the affine-invariant metric. In the case of a truly separable covariance, we demonstrate a joint approximation in the multiway space outperforms a mean-field approximation in optimization efficiency and provides a superior approximation to an unstructured Inverse-Wishart posterior under the average Mahalanobis distance of the data while maintaining a multiway interpretation. We moreover establish efficient expressions for the Euclidean and Riemannian gradients in both cases of the joint and mean-field approximation. We end with an analysis of commodity trade data.
title Geodesic Variational Bayes for Multiway Covariances
topic Computation
url https://arxiv.org/abs/2501.04935