Abstract:Uncertainty propagation in complex engineering systems often poses significant computational challenges related to modeling and quantifying probability distributions of model outputs, as those emerge as the result of various sources of uncertainty that are inherent in the system under investigation. Gaussian Processes regression (GPs) is a robust meta-modeling technique that allows for fast model prediction and exploration of response surfaces. Multi-fidelity variations of GPs further leverage information from cheap and low fidelity model simulations in order to improve their predictive performance on the high fidelity model. In order to cope with the high volume of data required to train GPs in high dimensional design spaces, a common practice is to introduce latent design variables that are typically projections of the original input space to a lower dimensional subspace, and therefore substitute the problem of learning the initial high dimensional mapping, with that of training a GP on a low dimensional space. In this paper, we present a Bayesian approach to identify optimal transformations that map the input points to low dimensional latent variables. The “projection” mapping consists of an orthonormal matrix that is considered a priori unknown and needs to be inferred jointly with the GP parameters, conditioned on the available training data. The proposed Bayesian inference scheme relies on a two-step iterative algorithm that samples from the marginal posteriors of the GP parameters and the projection matrix respectively, both using Markov Chain Monte Carlo (MCMC) sampling. In order to take into account the orthogonality constraints imposed on the orthonormal projection matrix, a Geodesic Monte Carlo sampling algorithm is employed, that is suitable for exploiting probability measures on manifolds. We extend the proposed framework to multi-fidelity models using GPs including the scenarios of training multiple outputs together. We validate our framework on three synthetic problems with a known lower-dimensional subspace. The benefits of our proposed framework, are illustrated on the computationally challenging aerodynamic optimization of a last-stage blade for an industrial gas turbine, where we study the effect of an 85-dimensional shape parameterization of a three-dimensional airfoil on two output quantities of interest, specifically on the aerodynamic efficiency and the degree of reaction.

A STRATEGY FOR ADAPTIVE SAMPLING OF MULTI-FIDELITY GAUSSIAN PROCESSES TO REDUCE PREDICTIVE UNCERTAINTY

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

Dimensionality Reduction for Multi-Fidelity Gaussian Processes Using Bayesian Adaptation

A Latent Variable Approach for Non-Hierarchical Multi-Fidelity Adaptive Sampling

An Adaptive Gaussian Process‐based Method for Efficient Bayesian Experimental Design in Groundwater Contaminant Source Identification Problems

Bayesian Learning of Orthogonal Embeddings for Multi-Fidelity Gaussian Processes

Multi-Fidelity Adaptive Sampling for Surrogate-Based Optimization and Uncertainty Quantification

Multi-Fidelity Surrogate Based on Single Linear Regression

Analytical uncertainty quantification approach based on adaptive generalized co‐Gaussian process model

Assessing the Performance of an Adaptive Multi-Fidelity Gaussian Process with Noisy Training Data: A Statistical Analysis

Scalable Fully Bayesian Gaussian Process Modeling and Calibration With Adaptive Sequential Monte Carlo for Industrial Applications

Gradient-enhanced deep Gaussian processes for multifidelity modelling

Sequential adaptive design for emulating costly computer codes

Nonparametric Uncertainty Representation Method with Different Insufficient Data from Two Sources

Adaptive reliability analysis for multi-fidelity models using a collective learning strategy

Multi-fidelity No-U-Turn Sampling

Overview of Gaussian process based multi-fidelity techniques with variable relationship between fidelities

Posterior sampling with Adaptive Gaussian Processes in Bayesian parameter identification

An efficient latent map multi-fidelity Kriging model and adaptive point-selected strategy for reliability analysis with time-consuming simulations