Emulators for Large-scale Computer Experiments with Quantitative and Qualitative Inputs

2026-06-01

AI Summary

This study addresses the challenge of constructing accurate and computationally efficient surrogate models for large-scale computer experiments involving both qualitative and quantitative inputs. To this end, the authors propose a scalable modeling framework that integrates additive Gaussian processes with the Vecchia approximation, introducing a novel covariance function tailored for mixed input types. The resulting approach substantially improves computational efficiency while preserving high predictive accuracy. Notably, this work presents the first Gaussian process surrogate model capable of effectively handling large-scale mixed-input scenarios, thereby extending the applicability of existing methods and achieving a favorable balance between accuracy and scalability.

Abstract

Computer experiments with both quantitative and qualitative inputs have become common across various areas. However, constructing accurate and computationally efficient emulators for such experiments at large scales remains a significant challenge. We propose a novel, scalable framework for emulating computer experiments with mixed inputs. Our approach is based on a new covariance function integrating additive Gaussian Processes (GPs) to handle the mixed inputs, with Vecchia approximation for scalability. We demonstrate that methods for large-scale computer experiments can be effectively extended when paired with our proposed modeling framework.

Problem

• computer experiments
• mixed inputs
• emulators
• scalability
• Gaussian Processes

Innovation

• mixed inputs
• additive Gaussian Processes
• Vecchia approximation
• scalable emulation
• covariance function