Gaussian Process Regression

Also known as GP

Overview

Gaussian process regression (GPR) places a probability distribution over functions, with kernels encoding assumptions such as smoothness, periodicity, or shared structure; conditioning on observations yields predictions and uncertainty, especially useful when data are scarce or decisions depend on confidence. Research develops richer models—including nonstationary, additive, multi-output, deep, and physics-informed GPs—and applies them to time series, spatial fields, dynamical systems, scientific surrogates, and Bayesian optimization. Because exact inference scales cubically with sample count, scalable approaches use inducing points and variational inference, nearest-neighbor or Vecchia approximations, sparse and low-rank kernels, and streaming or distributed updates. A central tension is balancing expressive models and calibrated uncertainty against computational cost and sensitivity to kernel and hyperparameter choices; reliable uncertainty under misspecification and scalable high-dimensional inference remain open challenges.

Latest papers

121 papers13 papers in the last four weeks, down 24% on the four weeks before. 0.1% of all new papers.

Feb 23, 2025

Optimal Kernel Learning for Gaussian Process Models with High-Dimensional Input

Gaussian process (GP) regression is a popular surrogate modeling tool for computer simulations in engineering and scientific domains. However, it often struggles with high computational costs and low prediction accuracy when the simulation involves too many input variables. For some simulation models, the outputs may only be significantly influenced by a small subset of the input variables, referred to as the active variables''. We propose an optimal kernel learning approach to identify these active variables, thereby overcoming GP model limitations and enhancing system understanding. Our method approximates the original GP model's covariance function through a convex combination of kernel functions, each utilizing low-dimensional subsets of input variables. Inspired by the Fedorov-Wynn algorithm from optimal design literature, we develop an optimal kernel learning algorithm to determine this approximation. We incorporate the effect heredity principle, a concept borrowed from the field of design and analysis of experiments'', to facilitate efficient computation. Through several examples, we demonstrate that the proposed method provides competitive prediction and variable selection, with favorable computational performance in the examples considered. It is an effective solution for interpreting the surrogate GP regression and simplifying the complex underlying system.