Projection Pursuit Gaussian Process Regression

TL;DR

This paper introduces a projection pursuit Gaussian process regression model that uses dimension expansion to better approximate complex functions in high-dimensional computer experiments, outperforming traditional models.

Contribution

It proposes a novel projection pursuit approach with dimension expansion for Gaussian process regression, enabling modeling of more complex functions in high dimensions.

Findings

Outperforms traditional Gaussian process models in simulations

Uses gradient descent for efficient model training

Enables approximation of complex functions through dimension expansion

Abstract

A primary goal of computer experiments is to reconstruct the function given by the computer code via scattered evaluations. Traditional isotropic Gaussian process models suffer from the curse of dimensionality, when the input dimension is relatively high given limited data points. Gaussian process models with additive correlation functions are scalable to dimensionality, but they are more restrictive as they only work for additive functions. In this work, we consider a projection pursuit model, in which the nonparametric part is driven by an additive Gaussian process regression. We choose the dimension of the additive function higher than the original input dimension, and call this strategy "dimension expansion". We show that dimension expansion can help approximate more complex functions. A gradient descent algorithm is proposed for model training based on the maximum likelihood estimation. Simulation studies show that the proposed method outperforms the traditional Gaussian process models. The Supplementary Materials are available online.