Vecchia-approximated Deep Gaussian Processes for Computer Experiments
From MaRDI portal
Publication:84900
Abstract: Deep Gaussian processes (DGPs) upgrade ordinary GPs through functional composition, in which intermediate GP layers warp the original inputs, providing flexibility to model non-stationary dynamics. Two DGP regimes have emerged in recent literature. A "big data" regime, prevalent in machine learning, favors approximate, optimization-based inference for fast, high-fidelity prediction. A "small data" regime, preferred for computer surrogate modeling, deploys posterior integration for enhanced uncertainty quantification (UQ). We aim to bridge this gap by expanding the capabilities of Bayesian DGP posterior inference through the incorporation of the Vecchia approximation, allowing linear computational scaling without compromising accuracy or UQ. We are motivated by surrogate modeling of simulation campaigns with upwards of 100,000 runs - a size too large for previous fully-Bayesian implementations - and demonstrate prediction and UQ superior to that of "big data" competitors. All methods are implemented in the "deepgp" package on CRAN.
Cites work
- A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code
- A case study competition among methods for analyzing large spatial data
- A general framework for Vecchia approximations of Gaussian processes
- A nonstationary space-time Gaussian process model for partially converged simulations
- A unifying view of sparse approximate Gaussian process regression
- Analyzing Nonstationary Spatial Data Using Piecewise Gaussian Processes
- Analyzing stochastic computer models: a review with opportunities
- Approximating Likelihoods for Large Spatial Data Sets
- Bayesian Inference for Non-Stationary Spatial Covariance Structure via Spatial Deformations
- Bayesian calibration of computer models. (With discussion)
- Bayesian treed Gaussian process models with an application to computer modeling
- Deep Compositional Spatial Models
- Deep Gaussian process autoencoders for novelty detection
- Deep Gaussian processes for calibration of computer models (with discussion)
- Dimension Reduction for Gaussian Process Emulation: An Application to the Influence of Bathymetry on Tsunami Heights
- Efficient Algorithms for Bayesian Nearest Neighbor Gaussian Processes
- Efficient emulators of computer experiments using compactly supported correlation functions, with an application to cosmology
- Efficient global optimization of expensive black-box functions
- Emulating satellite drag from large simulation experiments
- Gaussian Predictive Process Models for Large Spatial Data Sets
- Gaussian process learning via Fisher scoring of Vecchia's approximation
- How deep are deep Gaussian processes?
- Improving the performance of predictive process modeling for large datasets
- Interpolation of spatial data. Some theory for kriging
- Knot selection in sparse Gaussian processes with a variational objective function
- Locally induced Gaussian processes for large-scale simulation experiments
- Microcolony and biofilm formation as a survival strategy for bacteria
- Monte Carlo sampling methods using Markov chains and their applications
- Patchwork kriging for large-scale Gaussian process regression
- Practical Heteroscedastic Gaussian Process Modeling for Large Simulation Experiments
- RcppArmadillo: accelerating R with high-performance C++ linear algebra
- Scaled Vecchia approximation for fast computer-model emulation
- Sparse spectrum Gaussian process regression
- Strictly Proper Scoring Rules, Prediction, and Estimation
- Uncertainty quantification of a computer model for binary black hole formation
- VPS: VORONOI PIECEWISE SURROGATE MODELS FOR HIGH-DIMENSIONAL DATA FITTING
This page was built for publication: Vecchia-approximated Deep Gaussian Processes for Computer Experiments
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q84900)
