• Home
  • Search
  • MAGMA: inference and prediction using multi-task Gaussian processes with common mean
  • Open Access IconOpen Access
  • Cite Icon11
  • https://doi.org/10.1007/s10994-022-06172-1Copy DOI Icon

MAGMA: inference and prediction using multi-task Gaussian processes with common mean

Show More
  • Abstract
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

A novel multi-task Gaussian process (GP) framework is proposed, by using a common mean process for sharing information across tasks. In particular, we investigate the problem of time series forecasting, with the objective to improve multiple-step-ahead predictions. The common mean process is defined as a GP for which the hyper-posterior distribution is tractable. Therefore an EM algorithm is derived for handling both hyper-parameters optimisation and hyper-posterior computation. Unlike previous approaches in the literature, the model fully accounts for uncertainty and can handle irregular grids of observations while maintaining explicit formulations, by modelling the mean process in a unified GP framework. Predictive analytical equations are provided, integrating information shared across tasks through a relevant prior mean. This approach greatly improves the predictive performances, even far from observations, and may reduce significantly the computational complexity compared to traditional multi-task GP models. Our overall algorithm is called Magma (standing for Multi tAsk GPs with common MeAn). The quality of the mean process estimation, predictive performances, and comparisons to alternatives are assessed in various simulated scenarios and on real datasets.

Loading PDF

Similar Papers
  • Conference Article
  • Citations2

Information fusion in multi-task Gaussian process models

  • Sep 01, 2012
  • Shrihari Vasudevan +2
  • Conference Article
  • Citations22

Multi-task Gaussian process models for biomedical applications

  • Jun 01, 2014
  • Robert Durichen +4
  • Research Article
  • Citations15

Correcting missingness in passively-generated mobile data with Multi-Task Gaussian Processes

  • Mar 07, 2024
  • Transportation Research Part C: Emerging Technologies
  • Ekin Uğurel +5
  • Research Article

A multi-task Gaussian processes method for risk identification during cable-stayed bridge cable replacement

  • Mar 01, 2026
  • Results in Engineering
  • Beiyang Zhang +5
  • Preprint Article

MAGMA Gaussian Processes to complete Venusian atmospheric profiles from SOIR 

  • Jul 03, 2024
  • Simon Lejoly +4
  • Research Article
  • Citations7

Automated Multi-class Classification of Remotely Sensed Hyperspectral Imagery Via Gaussian Processes with a Non-stationary Covariance Function

  • Dec 04, 2015
  • Mathematical Geosciences
  • Anna Chlingaryan +3
  • Research Article
  • Citations5

A conditional one-output likelihood formulation for multitask Gaussian processes

  • Aug 27, 2022
  • Neurocomputing
  • Óscar García-Hinde +2
  • Research Article
  • Citations2

A Conditional One-Output Likelihood Formulation for Multitask Gaussian Processes

  • Jan 01, 2022
  • SSRN Electronic Journal
  • Óscar García-Hinde +2
  • Research Article
  • Citations164

Multitask Gaussian processes for multivariate physiological time-series analysis.

  • Jan 01, 2015
  • IEEE Transactions on Biomedical Engineering
  • Robert Durichen +4
  • Research Article
  • Citations4

Notice of Removal: Signal Extrapolation for Image and Video Error Concealment Using Gaussian Processes With Adaptive Nonstationary Kernels

  • Oct 01, 2012
  • IEEE Signal Processing Letters
  • Hadi Asheri +2
  • Research Article
  • Citations1

Prospective prediction of childhood body mass index trajectories using multi-task Gaussian processes

  • Nov 15, 2024
  • International Journal of Obesity
  • Arthur Leroy +12
  • Conference Article
  • Citations21

A near-to-far non-parametric learning approach for estimating traversability in deformable terrain

  • Nov 01, 2013
  • Ken Ho +2
  • Research Article
  • Citations9

Spatiotemporal Multitask Learning for 3-D Dynamic Field Modeling

  • Nov 01, 2019
  • IEEE Transactions on Automation Science and Engineering
  • Di Wang +3
  • Research Article
  • Citations216

A Multivariate Timeseries Modeling Approach to Severity of Illness Assessment and Forecasting in ICU with Sparse, Heterogeneous Clinical Data

  • Feb 10, 2015
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Marzyeh Ghassemi +6
  • Conference Article
  • Citations6

A New Approach to Building the Gaussian Process Model for Expensive Multi-objective Optimization

  • Aug 01, 2019
  • Jianping Luo +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.