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.

Scalable Gaussian Processes for Characterizing Multidimensional Change Surfaces

Change Surfaces for Expressive Multidimensional Changepoints and Counterfactual Prediction

Scalable Gaussian Processes for Data-Driven Design Using Big Data With Categorical Factors

Solving High Frequency and Multi-Scale PDEs with Gaussian Processes

ProSpar-GP: scalable Gaussian process modeling with massive non-stationary datasets

Iterative Methods for Full-Scale Gaussian Process Approximations for Large Spatial Data

Towards Scalable Gaussian Process Modeling

Scalable Multi-Temporal Remote Sensing Change Data Generation via Simulating Stochastic Change Process

Unsupervised Anomaly and Change Detection with Multivariate Gaussianization

Sparse Gaussian Processes with Spherical Harmonic Features Revisited

GigaGS: Scaling up Planar-Based 3D Gaussians for Large Scene Surface Reconstruction

Thoughts on Massively Scalable Gaussian Processes

Mixture of Gaussian Processes and its Applications

Making Multi-Axis Gaussian Graphical Models Scalable to Millions of Samples and Features

Adaptive and Temporally Consistent Gaussian Surfels for Multi-view Dynamic Reconstruction

Scalable mixed-domain Gaussian process modeling and model reduction for longitudinal data

Understanding and comparing scalable Gaussian process regression for big data

A Scalable Partitioned Approach to Model Massive Nonstationary Non-Gaussian Spatial Datasets

Highly Scalable Bayesian Geostatistical Modeling via Meshed Gaussian Processes on Partitioned Domains

Parallel cross-validation: A scalable fitting method for Gaussian process models

Interpretable and Scalable Graphical Models for Complex Spatio-temporal Processes