Projection predictive model selection for Gaussian processes

TL;DR

This paper introduces a projection-based variable selection method for Gaussian process models that enhances interpretability, reduces measurement costs, and speeds up predictions by focusing on predictive relevance.

Contribution

The paper presents a novel projection predictive approach for selecting variables in Gaussian processes, outperforming ARD in relevance assessment.

Findings

Improved variable relevance assessment over ARD

Enhanced model interpretability and explainability

Reduced computation time for predictions

Abstract

We propose a new method for simplification of Gaussian process (GP) models by projecting the information contained in the full encompassing model and selecting a reduced number of variables based on their predictive relevance. Our results on synthetic and real world datasets show that the proposed method improves the assessment of variable relevance compared to the automatic relevance determination (ARD) via the length-scale parameters. We expect the method to be useful for improving explainability of the models, reducing the future measurement costs and reducing the computation time for making new predictions.