Efficient Gaussian Process Modeling for Large Replicated and Censored Data in Industrial Research
Abstract
Gaussian process regression is widely used for modeling responses in both physical and computer experiments. In practice, data from physical experiments may include replications and be subject to censoring due to equipment limitations or contextual constraints, and such datasets can be large, posing a scalability challenge for standard Gaussian process inference. Motivated by a real-world industrial research study, we develop a Gaussian process modeling framework that efficiently analyzes large replicated and censored data. Our approach integrates a Vecchia approximation for computational scalability with a likelihood-based formulation that fully exploits replication and rigorously incorporates information from censored observations, analogous to exact simulation from truncated distributions. Through experiments on synthetic data and a real case study from industrial research, we demonstrate that the proposed method yields improved predictive accuracy and reliable uncertainty quantification compared to existing approaches, making it well suited for large-scale applications with structured replication and incomplete data.
Related
- Modeling inhomogeneous spatial point configurations with applications to replicated patterns in real-life waiting crowds
- PRESCCO: Efficient Prediction Intervals under a Right-Censored Covariate
- On The Relevance Of Using Temperature For Mechanical Fault Detection In Industrial Robots Based On Dynamic Modeling
- Efficient estimation methods for non-Gaussian regression models in continuous time
- Hold-Out Scoring for Efficient Gaussian DAG Learning
BibTeX
@misc{zhang2026efficient,
title = {{Efficient Gaussian Process Modeling for Large Replicated and Censored Data in Industrial Research}},
author = {Ying Zhang and Emily L. Kang and Shuo Li},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.05684},
eprint = {2611.05684},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.05684},
}
RIS
TY - UNPB
TI - Efficient Gaussian Process Modeling for Large Replicated and Censored Data in Industrial Research
AU - Ying Zhang
AU - Emily L. Kang
AU - Shuo Li
PY - 2026
DA - 2026/11/03
PB - arXiv
DO - 10.48550/arxiv.2611.05684
UR - https://arxiv.org/abs/2611.05684
AB - Gaussian process regression is widely used for modeling responses in both physical and computer experiments. In practice, data from physical experiments may include replications and be subject to censoring due to equipment limitations or contextual constraints, and such datasets can be large, posing a scalability challenge for standard Gaussian process inference. Motivated by a real-world industrial research study, we develop a Gaussian process modeling framework that efficiently analyzes large replicated and censored data. Our approach integrates a Vecchia approximation for computational scalability with a likelihood-based formulation that fully exploits replication and rigorously incorporates information from censored observations, analogous to exact simulation from truncated distributions. Through experiments on synthetic data and a real case study from industrial research, we demonstrate that the proposed method yields improved predictive accuracy and reliable uncertainty quantification compared to existing approaches, making it well suited for large-scale applications with structured replication and incomplete data.
ER -