Data-driven surrogates for high dimensional models using Gaussian process regression on the Grassmann manifold
Abstract
This paper introduces a surrogate modeling scheme based on Grassmannian manifold learning to be used for cost-efficient predictions of high-dimensional stochastic systems. The method exploits subspace-structured features of each solution by projecting it onto a Grassmann manifold. This point-wise linear dimensionality reduction harnesses the structural information to assess the similarity between solutions at different points in the input parameter space. The method utilizes a solution clustering approach in order to identify regions of the parameter space over which solutions are sufficiently similarly such that they can be interpolated on the Grassmannian. In this clustering, the reduced-order solutions are partitioned into disjoint clusters on the Grassmann manifold using the eigen-structure of properly defined Grassmannian kernels and, the Karcher mean of each cluster is estimated. Then, the points in each cluster are projected onto the tangent space with origin at the corresponding Karcher mean using the exponential mapping. For each cluster, a Gaussian process regression model is trained that maps the input parameters of the system to the reduced solution points of the corresponding cluster projected onto the tangent space. Using this Gaussian process model, the full-field solution can be efficiently predicted at any new point in the parameter space. In certainmore »
- Authors:
-
- Johns Hopkins Univ., Baltimore, MD (United States)
- Publication Date:
- Research Org.:
- Johns Hopkins Univ., Baltimore, MD (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- OSTI Identifier:
- 1853643
- Alternate Identifier(s):
- OSTI ID: 1638520
- Grant/Contract Number:
- SC0020428
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Computer Methods in Applied Mechanics and Engineering
- Additional Journal Information:
- Journal Volume: 370; Journal Issue: C; Journal ID: ISSN 0045-7825
- Publisher:
- Elsevier
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 42 ENGINEERING; 97 MATHEMATICS AND COMPUTING; Engineering; Mathematics; Mechanics; Grassmann manifold; Spectral clustering; Gaussian process regression; Machine learning; Nonlinear projection; Interpolation
Citation Formats
Giovanis, D. G., and Shields, M. D. Data-driven surrogates for high dimensional models using Gaussian process regression on the Grassmann manifold. United States: N. p., 2020.
Web. doi:10.1016/j.cma.2020.113269.
Giovanis, D. G., & Shields, M. D. Data-driven surrogates for high dimensional models using Gaussian process regression on the Grassmann manifold. United States. https://doi.org/10.1016/j.cma.2020.113269
Giovanis, D. G., and Shields, M. D. Wed .
"Data-driven surrogates for high dimensional models using Gaussian process regression on the Grassmann manifold". United States. https://doi.org/10.1016/j.cma.2020.113269. https://www.osti.gov/servlets/purl/1853643.
@article{osti_1853643,
title = {Data-driven surrogates for high dimensional models using Gaussian process regression on the Grassmann manifold},
author = {Giovanis, D. G. and Shields, M. D.},
abstractNote = {This paper introduces a surrogate modeling scheme based on Grassmannian manifold learning to be used for cost-efficient predictions of high-dimensional stochastic systems. The method exploits subspace-structured features of each solution by projecting it onto a Grassmann manifold. This point-wise linear dimensionality reduction harnesses the structural information to assess the similarity between solutions at different points in the input parameter space. The method utilizes a solution clustering approach in order to identify regions of the parameter space over which solutions are sufficiently similarly such that they can be interpolated on the Grassmannian. In this clustering, the reduced-order solutions are partitioned into disjoint clusters on the Grassmann manifold using the eigen-structure of properly defined Grassmannian kernels and, the Karcher mean of each cluster is estimated. Then, the points in each cluster are projected onto the tangent space with origin at the corresponding Karcher mean using the exponential mapping. For each cluster, a Gaussian process regression model is trained that maps the input parameters of the system to the reduced solution points of the corresponding cluster projected onto the tangent space. Using this Gaussian process model, the full-field solution can be efficiently predicted at any new point in the parameter space. In certain cases, the solution clusters will span disjoint regions of the parameter space. In such cases, for each of the solution clusters we utilize a second, density-based spatial clustering to group their corresponding input parameter points in the Euclidean space. The proposed method is applied to two numerical examples. Here, the first is a nonlinear stochastic ordinary differential equation with uncertain initial conditions where the surrogate is used to predict the time history solution. The second involves modeling of plastic deformation in a model amorphous solid using the Shear Transformation Zone theory of plasticity, where the proposed surrogate is used to predict the full strain field of a material specimen under large shear strains.},
doi = {10.1016/j.cma.2020.113269},
journal = {Computer Methods in Applied Mechanics and Engineering},
number = C,
volume = 370,
place = {United States},
year = {Wed Jul 15 00:00:00 EDT 2020},
month = {Wed Jul 15 00:00:00 EDT 2020}
}
Works referenced in this record:
Probabilistic learning for modeling and quantifying model-form uncertainties in nonlinear computational mechanics: PROBABILISTIC LEARNING FOR MODEL-FORM UNCERTAINTIES
journal, November 2018
- Soize, C.; Farhat, C.
- International Journal for Numerical Methods in Engineering, Vol. 117, Issue 7
Normalized cuts and image segmentation
journal, January 2000
- Jianbo Shi, ; Malik, J.
- IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, Issue 8
Accelerated scale bridging with sparsely approximated Gaussian learning
journal, February 2020
- Wang, Ting; Leiter, Kenneth W.; Plecháč, Petr
- Journal of Computational Physics, Vol. 403
The multi-element probabilistic collocation method (ME-PCM): Error analysis and applications
journal, November 2008
- Foo, Jasmine; Wan, Xiaoliang; Karniadakis, George Em
- Journal of Computational Physics, Vol. 227, Issue 22
A Global Geometric Framework for Nonlinear Dimensionality Reduction
journal, December 2000
- Tenenbaum, J. B.
- Science, Vol. 290, Issue 5500
Diffusion maps
journal, July 2006
- Coifman, Ronald R.; Lafon, Stéphane
- Applied and Computational Harmonic Analysis, Vol. 21, Issue 1
An Online Method for Interpolating Linear Parametric Reduced-Order Models
journal, January 2011
- Amsallem, David; Farhat, Charbel
- SIAM Journal on Scientific Computing, Vol. 33, Issue 5
Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
journal, February 2019
- Raissi, M.; Perdikaris, P.; Karniadakis, G. E.
- Journal of Computational Physics, Vol. 378
Bayesian identification of a projection-based reduced order model for computational fluid dynamics
journal, April 2020
- Stabile, Giovanni; Rosic, Bojana
- Computers & Fluids, Vol. 201
Polynomial-Chaos-Based Kriging
journal, January 2015
- Schobi, Roland; Sudret, Bruno; Wiart, Joe
- International Journal for Uncertainty Quantification, Vol. 5, Issue 2
Model Reduction for Large-Scale Systems with High-Dimensional Parametric Input Space
journal, January 2008
- Bui-Thanh, T.; Willcox, K.; Ghattas, O.
- SIAM Journal on Scientific Computing, Vol. 30, Issue 6
PLS-based adaptation for efficient PCE representation in high dimensions
journal, June 2019
- Papaioannou, Iason; Ehre, Max; Straub, Daniel
- Journal of Computational Physics, Vol. 387
Modeling and Quantification of Model-Form Uncertainties in Eigenvalue Computations Using a Stochastic Reduced Model
journal, March 2018
- Farhat, Charbel; Bos, Adrien; Avery, Philip
- AIAA Journal, Vol. 56, Issue 3
High-Order Collocation Methods for Differential Equations with Random Inputs
journal, January 2005
- Xiu, Dongbin; Hesthaven, Jan S.
- SIAM Journal on Scientific Computing, Vol. 27, Issue 3
Uncertainty Quantification in Multiscale Simulation of Materials: A Prospective
journal, July 2013
- Chernatynskiy, Aleksandr; Phillpot, Simon R.; LeSar, Richard
- Annual Review of Materials Research, Vol. 43, Issue 1
Nonlinear Component Analysis as a Kernel Eigenvalue Problem
journal, July 1998
- Schölkopf, Bernhard; Smola, Alexander; Müller, Klaus-Robert
- Neural Computation, Vol. 10, Issue 5
Dynamical Properties of Truncated Wiener-Hermite Expansions
journal, January 1967
- Orszag, Steven A.
- Physics of Fluids, Vol. 10, Issue 12
Adaptive sparse polynomial chaos expansion based on least angle regression
journal, March 2011
- Blatman, Géraud; Sudret, Bruno
- Journal of Computational Physics, Vol. 230, Issue 6
Multi-Element Generalized Polynomial Chaos for Arbitrary Probability Measures
journal, January 2006
- Wan, Xiaoliang; Karniadakis, George Em
- SIAM Journal on Scientific Computing, Vol. 28, Issue 3
Hessian eigenmaps: Locally linear embedding techniques for high-dimensional data
journal, April 2003
- Donoho, D. L.; Grimes, C.
- Proceedings of the National Academy of Sciences, Vol. 100, Issue 10
Modeling uncertainties in molecular dynamics simulations using a stochastic reduced-order basis
journal, September 2019
- Wang, Haoran; Guilleminot, Johann; Soize, Christian
- Computer Methods in Applied Mechanics and Engineering, Vol. 354
Nonlinear Dimensionality Reduction by Locally Linear Embedding
journal, December 2000
- Roweis, S. T.
- Science, Vol. 290, Issue 5500
Dynamics of viscoplastic deformation in amorphous solids
journal, June 1998
- Falk, M. L.; Langer, J. S.
- Physical Review E, Vol. 57, Issue 6
A Survey of Projection-Based Model Reduction Methods for Parametric Dynamical Systems
journal, January 2015
- Benner, Peter; Gugercin, Serkan; Willcox, Karen
- SIAM Review, Vol. 57, Issue 4
Deformation and Failure of Amorphous, Solidlike Materials
journal, March 2011
- Falk, Michael L.; Langer, J. S.
- Annual Review of Condensed Matter Physics, Vol. 2, Issue 1
Evaluation of the Disorder Temperature and Free-Volume Formalisms via Simulations of Shear Banding in Amorphous Solids
journal, May 2007
- Shi, Yunfeng; Katz, Michael B.; Li, Hui
- Physical Review Letters, Vol. 98, Issue 18
Principal Manifolds and Nonlinear Dimensionality Reduction via Tangent Space Alignment
journal, January 2004
- Zhang, Zhenyue; Zha, Hongyuan
- SIAM Journal on Scientific Computing, Vol. 26, Issue 1
Uncertainty quantification for complex systems with very high dimensional response using Grassmann manifold variations
journal, July 2018
- Giovanis, D. G.; Shields, M. D.
- Journal of Computational Physics, Vol. 364
Riemannian center of mass and mollifier smoothing
journal, September 1977
- Karcher, H.
- Communications on Pure and Applied Mathematics, Vol. 30, Issue 5
An adaptive hierarchical sparse grid collocation algorithm for the solution of stochastic differential equations
journal, May 2009
- Ma, Xiang; Zabaras, Nicholas
- Journal of Computational Physics, Vol. 228, Issue 8
Interpolation Method for Adapting Reduced-Order Models and Application to Aeroelasticity
journal, July 2008
- Amsallem, David; Farhat, Charbel
- AIAA Journal, Vol. 46, Issue 7
Data-driven probability concentration and sampling on manifold
journal, September 2016
- Soize, C.; Ghanem, R.
- Journal of Computational Physics, Vol. 321
Sparse On-Line Gaussian Processes
journal, March 2002
- Csató, Lehel; Opper, Manfred
- Neural Computation, Vol. 14, Issue 3
Reduced Basis Approximation and a Posteriori Error Estimation for Affinely Parametrized Elliptic Coercive Partial Differential Equations: Application to Transport and Continuum Mechanics
journal, May 2008
- Rozza, G.; Huynh, D. B. P.; Patera, A. T.
- Archives of Computational Methods in Engineering, Vol. 15, Issue 3
AK-MCS: An active learning reliability method combining Kriging and Monte Carlo Simulation
journal, March 2011
- Echard, B.; Gayton, N.; Lemaire, M.
- Structural Safety, Vol. 33, Issue 2
The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
journal, January 2002
- Xiu, Dongbin; Karniadakis, George Em
- SIAM Journal on Scientific Computing, Vol. 24, Issue 2
Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
journal, June 2003
- Belkin, Mikhail; Niyogi, Partha
- Neural Computation, Vol. 15, Issue 6
Variance‐based simplex stochastic collocation with model order reduction for high‐dimensional systems
journal, November 2018
- Giovanis, D. G.; Shields, M. D.
- International Journal for Numerical Methods in Engineering, Vol. 117, Issue 11
Schubert Varieties and Distances between Subspaces of Different Dimensions
journal, January 2016
- Ye, Ke; Lim, Lek-Heng
- SIAM Journal on Matrix Analysis and Applications, Vol. 37, Issue 3
A nonparametric probabilistic approach for quantifying uncertainties in low-dimensional and high-dimensional nonlinear models: NONPARAMETRIC PROBABILISTIC APPROACH FOR UQ IN NONLINEAR MODELS
journal, June 2016
- Soize, C.; Farhat, C.
- International Journal for Numerical Methods in Engineering, Vol. 109, Issue 6
An adaptive multi-element generalized polynomial chaos method for stochastic differential equations
journal, November 2005
- Wan, Xiaoliang; Karniadakis, George Em
- Journal of Computational Physics, Vol. 209, Issue 2
A tutorial on spectral clustering
journal, August 2007
- von Luxburg, Ulrike
- Statistics and Computing, Vol. 17, Issue 4
A Stochastic Collocation Method for Elliptic Partial Differential Equations with Random Input Data
journal, January 2007
- Babuška, Ivo; Nobile, Fabio; Tempone, Raúl
- SIAM Journal on Numerical Analysis, Vol. 45, Issue 3
Extending Classical Surrogate Modeling to high Dimensions Through Supervised Dimensionality Reduction: a Data-Driven Approach
journal, January 2020
- Lataniotis, Christos; Marelli, Stefano; Sudret, Bruno
- International Journal for Uncertainty Quantification, Vol. 10, Issue 1
Parallel three-dimensional simulations of quasi-static elastoplastic solids
journal, March 2020
- Boffi, Nicholas M.; Rycroft, Chris H.
- Computer Physics Communications
The Geometry of Algorithms with Orthogonality Constraints
journal, January 1998
- Edelman, Alan; Arias, Tomás A.; Smith, Steven T.
- SIAM Journal on Matrix Analysis and Applications, Vol. 20, Issue 2
Riemannian Geometry of Grassmann Manifolds with a View on Algorithmic Computation
journal, January 2004
- Absil, P. -A.; Mahony, R.; Sepulchre, R.
- Acta Applicandae Mathematicae, Vol. 80, Issue 2
An adaptive local reduced basis method for solving PDEs with uncertain inputs and evaluating risk
journal, March 2019
- Zou, Zilong; Kouri, Drew; Aquino, Wilkins
- Computer Methods in Applied Mechanics and Engineering, Vol. 345
An adaptive hierarchical sparse grid collocation algorithm for the solution of stochastic differential equations
journal, May 2009
- Ma, Xiang; Zabaras, Nicholas
- Journal of Computational Physics, Vol. 228, Issue 8
Adaptive sparse polynomial chaos expansion based on least angle regression
journal, March 2011
- Blatman, Géraud; Sudret, Bruno
- Journal of Computational Physics, Vol. 230, Issue 6
Global sensitivity analysis using polynomial chaos expansions
journal, July 2008
- Sudret, Bruno
- Reliability Engineering & System Safety, Vol. 93, Issue 7
Dynamical Properties of Truncated Wiener-Hermite Expansions
journal, January 1967
- Orszag, Steven A.
- Physics of Fluids, Vol. 10, Issue 12
Nonequilibrium thermodynamics of driven amorphous materials. I. Internal degrees of freedom and volume deformation
journal, September 2009
- Bouchbinder, Eran; Langer, J. S.
- Physical Review E, Vol. 80, Issue 3
Coarse graining atomistic simulations of plastically deforming amorphous solids
journal, May 2017
- Hinkle, Adam R.; Rycroft, Chris H.; Shields, Michael D.
- Physical Review E, Vol. 95, Issue 5
Nonlinear Dimensionality Reduction by Locally Linear Embedding
journal, December 2000
- Roweis, S. T.
- Science, Vol. 290, Issue 5500
High-Order Collocation Methods for Differential Equations with Random Inputs
journal, January 2005
- Xiu, Dongbin; Hesthaven, Jan S.
- SIAM Journal on Scientific Computing, Vol. 27, Issue 3
Sparse On-Line Gaussian Processes
journal, March 2002
- Csató, Lehel; Opper, Manfred
- Neural Computation, Vol. 14, Issue 3
Nonequilibrium Thermodynamics of Amorphous Materials II: Effective-Temperature Theory
text, January 2009
- Bouchbinder, Eran; Langer, J. S.
- arXiv
Nonequilibrium Thermodynamics of Amorphous Materials III: Shear-Transformation-Zone Plasticity
text, January 2009
- Bouchbinder, Eran; Langer, J. S.
- arXiv
Uncertainty quantification for complex systems with very high dimensional response using Grassmann manifold variations
text, January 2017
- Giovanis, D. G.; Shields, M. D.
- arXiv
Works referencing / citing this record:
Grassmannian diffusion maps based dimension reduction and classification for high-dimensional data
preprint, January 2020
- Santos, K. R. M. dos; Giovanis, D. G.; Shields, M. D.
- arXiv