A Generalized Bias-Variance Decomposition for Bregman Divergences

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autore principale: Pfau, David
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914152748941312
author Pfau, David
author_facet Pfau, David
contents The bias-variance decomposition is a central result in statistics and machine learning, but is typically presented only for the squared error. We present a generalization of the bias-variance decomposition where the prediction error is a Bregman divergence, which is relevant to maximum likelihood estimation with exponential families. While the result is already known, there was not previously a clear, standalone derivation, so we provide one for pedagogical purposes. A version of this note previously appeared on the author's personal website without context. Here we provide additional discussion and references to the relevant prior literature.
format Preprint
id arxiv_https___arxiv_org_abs_2511_08789
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Generalized Bias-Variance Decomposition for Bregman Divergences
Pfau, David
Machine Learning
The bias-variance decomposition is a central result in statistics and machine learning, but is typically presented only for the squared error. We present a generalization of the bias-variance decomposition where the prediction error is a Bregman divergence, which is relevant to maximum likelihood estimation with exponential families. While the result is already known, there was not previously a clear, standalone derivation, so we provide one for pedagogical purposes. A version of this note previously appeared on the author's personal website without context. Here we provide additional discussion and references to the relevant prior literature.
title A Generalized Bias-Variance Decomposition for Bregman Divergences
topic Machine Learning
url https://arxiv.org/abs/2511.08789