Robust, Online, and Adaptive Decentralized Gaussian Processes

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Llorente, Fernando, Waxman, Daniel, Jantre, Sanket, Urban, Nathan M., Minkoff, Susan E.
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912599683104768
author Llorente, Fernando
Waxman, Daniel
Jantre, Sanket
Urban, Nathan M.
Minkoff, Susan E.
author_facet Llorente, Fernando
Waxman, Daniel
Jantre, Sanket
Urban, Nathan M.
Minkoff, Susan E.
contents Gaussian processes (GPs) offer a flexible, uncertainty-aware framework for modeling complex signals, but scale cubically with data, assume static targets, and are brittle to outliers, limiting their applicability in large-scale problems with dynamic and noisy environments. Recent work introduced decentralized random Fourier feature Gaussian processes (DRFGP), an online and distributed algorithm that casts GPs in an information-filter form, enabling exact sequential inference and fully distributed computation without reliance on a fusion center. In this paper, we extend DRFGP along two key directions: first, by introducing a robust-filtering update that downweights the impact of atypical observations; and second, by incorporating a dynamic adaptation mechanism that adapts to time-varying functions. The resulting algorithm retains the recursive information-filter structure while enhancing stability and accuracy. We demonstrate its effectiveness on a large-scale Earth system application, underscoring its potential for in-situ modeling.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18011
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Robust, Online, and Adaptive Decentralized Gaussian Processes
Llorente, Fernando
Waxman, Daniel
Jantre, Sanket
Urban, Nathan M.
Minkoff, Susan E.
Machine Learning
Multiagent Systems
Signal Processing
Gaussian processes (GPs) offer a flexible, uncertainty-aware framework for modeling complex signals, but scale cubically with data, assume static targets, and are brittle to outliers, limiting their applicability in large-scale problems with dynamic and noisy environments. Recent work introduced decentralized random Fourier feature Gaussian processes (DRFGP), an online and distributed algorithm that casts GPs in an information-filter form, enabling exact sequential inference and fully distributed computation without reliance on a fusion center. In this paper, we extend DRFGP along two key directions: first, by introducing a robust-filtering update that downweights the impact of atypical observations; and second, by incorporating a dynamic adaptation mechanism that adapts to time-varying functions. The resulting algorithm retains the recursive information-filter structure while enhancing stability and accuracy. We demonstrate its effectiveness on a large-scale Earth system application, underscoring its potential for in-situ modeling.
title Robust, Online, and Adaptive Decentralized Gaussian Processes
topic Machine Learning
Multiagent Systems
Signal Processing
url https://arxiv.org/abs/2509.18011