English

Federated Gaussian Process: Convergence, Automatic Personalization and Multi-fidelity Modeling

Machine Learning 2024-04-01 v1 Machine Learning

Abstract

In this paper, we propose \texttt{FGPR}: a Federated Gaussian process (GP\mathcal{GP}) regression framework that uses an averaging strategy for model aggregation and stochastic gradient descent for local client computations. Notably, the resulting global model excels in personalization as \texttt{FGPR} jointly learns a global GP\mathcal{GP} prior across all clients. The predictive posterior then is obtained by exploiting this prior and conditioning on local data which encodes personalized features from a specific client. Theoretically, we show that \texttt{FGPR} converges to a critical point of the full log-likelihood function, subject to statistical error. Through extensive case studies we show that \texttt{FGPR} excels in a wide range of applications and is a promising approach for privacy-preserving multi-fidelity data modeling.

Keywords

Cite

@article{arxiv.2111.14008,
  title  = {Federated Gaussian Process: Convergence, Automatic Personalization and Multi-fidelity Modeling},
  author = {Xubo Yue and Raed Al Kontar},
  journal= {arXiv preprint arXiv:2111.14008},
  year   = {2024}
}

Comments

56 pages. Under Review

R2 v1 2026-06-24T07:54:23.502Z