Statistical deep learning for spatial and spatiotemporal data

CK Wikle, A Zammit-Mangion - Annual Review of Statistics and …, 2023 - annualreviews.org
Deep neural network models have become ubiquitous in recent years and have been
applied to nearly all areas of science, engineering, and industry. These models are …

Statistical deep learning for spatial and spatio-temporal data

CK Wikle, A Zammit-Mangion - arXiv preprint arXiv:2206.02218, 2022 - arxiv.org
Deep neural network models have become ubiquitous in recent years, and have been
applied to nearly all areas of science, engineering, and industry. These models are …

Variational sparse inverse Cholesky approximation for latent Gaussian processes via double Kullback-Leibler minimization

J Cao, M Kang, F Jimenez, H Sang… - International …, 2023 - proceedings.mlr.press
To achieve scalable and accurate inference for latent Gaussian processes, we propose a
variational approximation based on a family of Gaussian distributions whose covariance …

Non-stationary Gaussian process surrogates

A Sauer, A Cooper, RB Gramacy - arXiv preprint arXiv:2305.19242, 2023 - arxiv.org
We provide a survey of non-stationary surrogate models which utilize Gaussian processes
(GPs) or variations thereof, including non-stationary kernel adaptations, partition and local …

Traditional kriging versus modern Gaussian processes for large‐scale mining data

RB Christianson, RM Pollyea… - Statistical Analysis and …, 2023 - Wiley Online Library
The canonical technique for nonlinear modeling of spatial/point‐referenced data is known
as kriging in geostatistics, and as Gaussian Process (GP) regression for surrogate modeling …

Free-form variational inference for Gaussian process state-space models

X Fan, EV Bonilla, T O'Kane… - … Conference on Machine …, 2023 - proceedings.mlr.press
Gaussian process state-space models (GPSSMs) provide a principled and flexible approach
to modeling the dynamics of a latent state, which is observed at discrete-time points via a …

Optimizing cervical cancer classification using transfer learning with deep gaussian processes and support vector machines

E Ahishakiye, F Kanobe - Discover Artificial Intelligence, 2024 - Springer
Background Cervical cancer is the fourth most frequent cancer in women worldwide. Even
though cervical cancer deaths have decreased significantly in Western countries, low and …

[HTML][HTML] SECRET: Statistical Emulation for Computational Reverse Engineering and Translation with applications in healthcare

LM Paun, MJ Colebank, A Taylor-LaPole… - Computer Methods in …, 2024 - Elsevier
There have been impressive advances in the physical and mathematical modelling of
complex physiological systems in the last few decades, with the potential to revolutionise …

Contour Location for Reliability in Airfoil Simulation Experiments using Deep Gaussian Processes

AS Booth, SA Renganathan, RB Gramacy - arXiv preprint arXiv …, 2023 - arxiv.org
Bayesian deep Gaussian processes (DGPs) outperform ordinary GPs as surrogate models
of complex computer experiments when response surface dynamics are non-stationary …

Robust expected improvement for Bayesian optimization

RB Christianson, RB Gramacy - IISE Transactions, 2024 - Taylor & Francis
Abstract Bayesian Optimization (BO) links Gaussian Process (GP) surrogates with
sequential design toward optimizing expensive-to-evaluate black-box functions. Example …