Salvato in:
Dettagli Bibliografici
Autori principali: Jacobsen, Andrew, Cutkosky, Ashok
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2405.19175
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910462869766144
author Jacobsen, Andrew
Cutkosky, Ashok
author_facet Jacobsen, Andrew
Cutkosky, Ashok
contents We develop algorithms for online linear regression which achieve optimal static and dynamic regret guarantees \emph{even in the complete absence of prior knowledge}. We present a novel analysis showing that a discounted variant of the Vovk-Azoury-Warmuth forecaster achieves dynamic regret of the form $R_{T}(\vec{u})\le O\left(d\log(T)\vee \sqrt{dP_{T}^γ(\vec{u})T}\right)$, where $P_{T}^γ(\vec{u})$ is a measure of variability of the comparator sequence, and show that the discount factor achieving this result can be learned on-the-fly. We show that this result is optimal by providing a matching lower bound. We also extend our results to \emph{strongly-adaptive} guarantees which hold over every sub-interval $[a,b]\subseteq[1,T]$ simultaneously.
format Preprint
id arxiv_https___arxiv_org_abs_2405_19175
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Online Linear Regression in Dynamic Environments via Discounting
Jacobsen, Andrew
Cutkosky, Ashok
Machine Learning
We develop algorithms for online linear regression which achieve optimal static and dynamic regret guarantees \emph{even in the complete absence of prior knowledge}. We present a novel analysis showing that a discounted variant of the Vovk-Azoury-Warmuth forecaster achieves dynamic regret of the form $R_{T}(\vec{u})\le O\left(d\log(T)\vee \sqrt{dP_{T}^γ(\vec{u})T}\right)$, where $P_{T}^γ(\vec{u})$ is a measure of variability of the comparator sequence, and show that the discount factor achieving this result can be learned on-the-fly. We show that this result is optimal by providing a matching lower bound. We also extend our results to \emph{strongly-adaptive} guarantees which hold over every sub-interval $[a,b]\subseteq[1,T]$ simultaneously.
title Online Linear Regression in Dynamic Environments via Discounting
topic Machine Learning
url https://arxiv.org/abs/2405.19175