November 30, 2025EnvironmetricsOpen Access

Compactly‐Supported Nonstationary Kernels for Computing Exact Gaussian Processes on Big Data

View Full Paper

Ask AI

Bookmark

Authors

MRMark D. Risser MNMarcus M. Noack HLHengrui Luo

Discussion

Member takes

Implication

Development of a Bayesian model with nonstationary kernels enhances performance over 1 million temperature measurements, indicating strength against machine learning methods.

Key Points

Massive data set analysis yields improved performance using a novel Bayesian model with nonstationary kernels, outperforming existing methods.
Experimental results demonstrate enhancements in accuracy when applying these kernels to temperature data from over 1 million points.
Assessment of synthetic data showcases favorable performance in the Gaussian process framework with new kernels and high-performance computing support.
Implications suggest that scalable, exact GPs could compete with prevalent machine learning approaches in extensive data applications.

Cite This Study

Risser et al. (2025) studied this question.

synapsesocial.com/papers/692b94261d383f2b2a37837a https://doi.org/10.1002/env.70054

View Full Paper

Ask AI

Bookmark

Implication

Development of a Bayesian model with nonstationary kernels enhances performance over 1 million temperature measurements, indicating strength against machine learning methods.

Compactly‐Supported Nonstationary Kernels for Computing Exact Gaussian Processes on Big Data

Discussion

Member takes

Implication

Key Points

Cite This Study

Compactly‐Supported Nonstationary Kernels for Computing Exact Gaussian Processes on Big Data

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider