Accelerated scale bridging with sparsely approximated Gaussian learning
From MaRDI portal
Publication:2222967
DOI10.1016/J.JCP.2019.109049zbMATH Open1453.62650arXiv1901.06777OpenAlexW2982104475WikidataQ126991956 ScholiaQ126991956MaRDI QIDQ2222967FDOQ2222967
Authors: Ting Wang, K. Leiter, Petr Plecháč, Jaroslaw Knap 
Publication date: 28 January 2021
Published in: Journal of Computational Physics (Search for Journal in Brave)
Abstract: Multiscale modeling is a systematic approach to describe the behavior of complex systems by coupling models from different scales. The approach has been demonstrated to be very effective in areas of science as diverse as materials science, climate modeling and chemistry. However, routine use of multiscale simulations is often hindered by the very high cost of individual at-scale models. Approaches aiming to alleviate that cost by means of Gaussian process regression based surrogate models have been proposed. Yet, many of these surrogate models are expensive to construct, especially when the number of data needed is large. In this article, we employ a hierarchical sparse Cholesky decomposition to develop a sparse Gaussian process regression method and apply the method to approximate the equation of state of an energetic material in a multiscale model of dynamic deformation. We demonstrate that the method provides a substantial reduction both in computational cost and solution error as compared with previous methods.
Full work available at URL: https://arxiv.org/abs/1901.06777
Recommendations
- Surrogate modeling of multiscale models using kernel methods
- Multiresolution clustering analysis for efficient modeling of hierarchical material systems
- Evaluation of convergence behavior of metamodeling techniques for bridging scales in multi-scale multimaterial simulation
- Sequential multiscale modeling using sparse representation
- Multiscale materials
Cites Work
- Multigrid with Rough Coefficients and Multiresolution Operator Decomposition from Hierarchical Information Games
- A unifying view of sparse approximate Gaussian process regression
- Gaussian processes with built-in dimensionality reduction: applications to high-dimensional uncertainty propagation
- Mixed finite element methods and applications
- Bayesian numerical homogenization
- An improved algorithm for \textit{in situ} adaptive tabulation
- Evaluation of convergence behavior of metamodeling techniques for bridging scales in multi-scale multimaterial simulation
- Computationally efficient implementation of combustion chemistry usingin situadaptive tabulation
- Adaptive sampling in hierarchical simulation
- Advanced Lectures on Machine Learning
- Surrogate-based methods
- A call to arms for task parallelism in multi-scale materials modeling
- Active learning of constitutive relation from mesoscopic dynamics for macroscopic modeling of non-Newtonian flows
- Spatial adaptive sampling in multiscale simulation
- High-resolution coupled physics solvers for analysing fine-scale nuclear reactor design problems
Cited In (5)
- Sequential multiscale modeling using sparse representation
- Data-driven surrogates for high dimensional models using Gaussian process regression on the Grassmann manifold
- Filters for improvement of multiscale data from atomistic simulations
- Efficient hybrid explicit-implicit learning for multiscale problems
- Temporal scale-bridging of chemistry in a multiscale model: application to reactivity of an energetic material
Uses Software
This page was built for publication: Accelerated scale bridging with sparsely approximated Gaussian learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2222967)
