A graphical Gaussian process model for multi-fidelity emulation of expensive computer codes

07/31/2021

∙

We present a novel Graphical Multi-fidelity Gaussian Process (GMGP) model that uses a directed acyclic graph to model dependencies between multi-fidelity simulation codes. The proposed model is an extension of the Kennedy-O'Hagan model for problems where different codes cannot be ranked in a sequence from lowest to highest fidelity.

READ FULL TEXT

A graphical Gaussian process model for multi-fidelity emulation of expensive computer codes

Multi-fidelity Gaussian Process for Biomanufacturing Process Modeling with Small Data

Multi-Stage Multi-Fidelity Gaussian Process Modeling, with Application to Heavy-Ion Collisions

Residual Gaussian Process: A Tractable Nonparametric Bayesian Emulator for Multi-fidelity Simulations

Stacking designs: designing multi-fidelity computer experiments with confidence

Laplace Inference for Multi-fidelity Gaussian Process Classification

Multi-fidelity classification using Gaussian processes: accelerating the prediction of large-scale computational models

Multi-fidelity Gaussian Process Bandit Optimisation

A graphical Gaussian process model for multi-fidelity emulation of expensive computer codes

Related Research

Multi-fidelity Gaussian Process for Biomanufacturing Process Modeling with Small Data

Multi-Stage Multi-Fidelity Gaussian Process Modeling, with Application to Heavy-Ion Collisions

Residual Gaussian Process: A Tractable Nonparametric Bayesian Emulator for Multi-fidelity Simulations

Stacking designs: designing multi-fidelity computer experiments with confidence

Laplace Inference for Multi-fidelity Gaussian Process Classification

Multi-fidelity classification using Gaussian processes: accelerating the prediction of large-scale computational models

Multi-fidelity Gaussian Process Bandit Optimisation