Multi-fidelity modeling with different input domain definitions using Deep Gaussian Processes

06/29/2020
by   Ali Hebbal, et al.
0

Multi-fidelity approaches combine different models built on a scarce but accurate data-set (high-fidelity data-set), and a large but approximate one (low-fidelity data-set) in order to improve the prediction accuracy. Gaussian Processes (GPs) are one of the popular approaches to exhibit the correlations between these different fidelity levels. Deep Gaussian Processes (DGPs) that are functional compositions of GPs have also been adapted to multi-fidelity using the Multi-Fidelity Deep Gaussian process model (MF-DGP). This model increases the expressive power compared to GPs by considering non-linear correlations between fidelities within a Bayesian framework. However, these multi-fidelity methods consider only the case where the inputs of the different fidelity models are defined over the same domain of definition (e.g., same variables, same dimensions). However, due to simplification in the modeling of the low-fidelity, some variables may be omitted or a different parametrization may be used compared to the high-fidelity model. In this paper, Deep Gaussian Processes for multi-fidelity (MF-DGP) are extended to the case where a different parametrization is used for each fidelity. The performance of the proposed multifidelity modeling technique is assessed on analytical test cases and on structural and aerodynamic real physical problems.

READ FULL TEXT

page 2

page 14

page 15

page 16

page 17

research
12/16/2018

Linking Gaussian Process regression with data-driven manifold embeddings for nonlinear data fusion

In statistical modeling with Gaussian Process regression, it has been sh...
research
04/26/2016

Deep Multi-fidelity Gaussian Processes

We develop a novel multi-fidelity framework that goes far beyond the cla...
research
06/30/2020

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

The design process of complex systems such as new configurations of airc...
research
05/09/2019

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

Machine learning techniques typically rely on large datasets to create a...
research
10/15/2019

Identification of Behavioural Models for Railway Turnouts Monitoring

This study introduces a low-complexity behavioural model to describe the...
research
11/10/2021

Safe Real-Time Optimization using Multi-Fidelity Gaussian Processes

This paper proposes a new class of real-time optimization schemes to ove...
research
10/16/2020

Multi-fidelity data fusion for the approximation of scalar functions with low intrinsic dimensionality through active subspaces

Gaussian processes are employed for non-parametric regression in a Bayes...

Please sign up or login with your details

Forgot password? Click here to reset