Abstract:Numerous engineering problems of interest to the industry are often characterized by expensive black-box objective experiments or computer simulations. Obtaining insight into the problem or performing subsequent optimizations requires hundreds of thousands of evaluations of the objective function which is most often a practically unachievable task. Gaussian Process (GP) surrogate modeling replaces the expensive function with a cheap-to-evaluate data-driven probabilistic model. While the GP does not assume a functional form of the problem, it is defined by a set of parameters, called hyperparameters. The hyperparameters define the characteristics of the objective function, such as smoothness, magnitude, periodicity, etc. Accurately estimating these hyperparameters is a key ingredient in developing a reliable and generalizable surrogate model. Markov chain Monte Carlo (MCMC) is a ubiquitously used Bayesian method to estimate these hyperparameters. At the GE Global Research Center, a customized industry-strength Bayesian hybrid modeling framework utilizing the GP, called GEBHM, has been employed and validated over many years. GEBHM is very effective on problems of small and medium size, typically less than 1000 training points. However, the GP does not scale well in time with a growing dataset and problem dimensionality which can be a major impediment in such problems. In this work, we extend and implement in GEBHM an Adaptive Sequential Monte Carlo (ASMC) methodology for training the GP enabling the modeling of large-scale industry problems. This implementation saves computational time (especially for large-scale problems) while not sacrificing predictability over the current MCMC implementation. We demonstrate the effectiveness and accuracy of GEBHM with ASMC on four mathematical problems and on two challenging industry applications of varying complexity.

Additive Gaussian process regression for metamodeling in high-dimensional problems

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

Additive Kernels for Gaussian Process Modeling

Towards Scalable Gaussian Process Modeling

Dimensionality Reduction for Multi-Fidelity Gaussian Processes Using Bayesian Adaptation

High-dimensional mixed-categorical Gaussian processes with application to multidisciplinary design optimization for a green aircraft

Remarks for Scaling Up a General Gaussian Process to Model Large Dataset with Sub-models

Data-Driven Design via Scalable Gaussian Processes for Multi-Response Big Data With Qualitative Factors

Latent Map Gaussian Processes for Mixed Variable Metamodeling

ANOVA Gaussian process modeling for high-dimensional stochastic computational models

Gaussian Process Neural Additive Models

Scalable Gaussian Process Using Inexact Admm For Big Data

A Latent Variable Approach to Gaussian Process Modeling with Qualitative and Quantitative Factors.

Gaussian Process Emulation For Big Data In Data-Driven Metamaterials Design

ANOVA-GP Modeling for High-Dimensional Bayesian Inverse Problems

Globally Approximate Gaussian Processes for Big Data with Application to Data-Driven Metamaterials Design (IDETC2019-98027)

Scaling Gaussian Process Regression with Derivatives

Leveraging the Nugget Parameter for Efficient Gaussian Process Modeling

Gaussian Process Regression with Soft Inequality and Monotonicity Constraints

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