MFNets: data efficient all-at-once learning of multifidelity surrogates as directed networks of information sources

Gorodetsky, Alex A.; Jakeman, John D.; Geraci, Gianluca

doi:10.1007/s00466-021-02042-0

MFNets: data efficient all-at-once learning of multifidelity surrogates as directed networks of information sources

Journal Article · 17 August 2021 · Computational Mechanics

DOI:https://doi.org/10.1007/s00466-021-02042-0· OSTI ID:1820427

^[1]; ^[2]; Geraci, Gianluca ^[2]

Univ. of Michigan, Ann Arbor, MI (United States)
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)

We present an approach for constructing a surrogate from ensembles of information sources of varying cost and accuracy. The multifidelity surrogate encodes connections between information sources as a directed acyclic graph, and is trained via gradient-based minimization of a nonlinear least squares objective. While the vast majority of state-of-the-art assumes hierarchical connections between information sources, our approach works with flexibly structured information sources that may not admit a strict hierarchy. The formulation has two advantages: (1) increased data efficiency due to parsimonious multifidelity networks that can be tailored to the application; and (2) no constraints on the training data—we can combine noisy, non-nested evaluations of the information sources. Finally, numerical examples ranging from synthetic to physics-based computational mechanics simulations indicate the error in our approach can be orders-of-magnitude smaller, particularly in the low-data regime, than single-fidelity and hierarchical multifidelity approaches.

View Accepted Manuscript (DOE)

Research Organization:: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)

Sponsoring Organization:: USDOE Laboratory Directed Research and Development (LDRD) Program; USDOE National Nuclear Security Administration (NNSA)

Grant/Contract Number:: NA0003525

OSTI ID:: 1820427

Report Number(s):: SAND--2021-10721J; 699125

Journal Information:: Computational Mechanics, Journal Name: Computational Mechanics Journal Issue: 4 Vol. 68; ISSN 0178-7675

Publisher:: SpringerCopyright Statement

Country of Publication:: United States

Language:: English

References (27)

Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling Perdikaris, P.; Raissi, M.; Damianou, A. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 473, Issue 2198 https://doi.org/10.1098/rspa.2016.0751	journal	February 2017
TIEGCM results associated with "Dynamics and electrodynamics of an UFKW packet in the ionosphere and thermosphere" Astrid, Maute, UCAR/NCAR - GDEX https://doi.org/10.5065/4382-8n24	dataset	January 2020
Adaptive multi‐index collocation for uncertainty quantification and sensitivity analysis Jakeman, John D.; Eldred, Michael S.; Geraci, Gianluca International Journal for Numerical Methods in Engineering, Vol. 121, Issue 6 https://doi.org/10.1002/nme.6268	journal	November 2019
A trust-region framework for managing the use of approximation models in optimization Alexandrov, N. M.; Dennis, J. E.; Lewis, R. M. Structural Optimization, Vol. 15, Issue 1 https://doi.org/10.1007/BF01197433	journal	February 1998
Multi-index Monte Carlo: when sparsity meets sampling Haji-Ali, Abdul-Lateef; Nobile, Fabio; Tempone, Raúl Numerische Mathematik, Vol. 132, Issue 4 https://doi.org/10.1007/s00211-015-0734-5	journal	June 2015
Coupling multi-fidelity kriging and model-order reduction for the construction of virtual charts Nachar, Stéphane; Boucard, Pierre-Alain; Néron, David Computational Mechanics, Vol. 64, Issue 6 https://doi.org/10.1007/s00466-019-01745-9	journal	June 2019
Bi-fidelity stochastic gradient descent for structural optimization under uncertainty De, Subhayan; Maute, Kurt; Doostan, Alireza Computational Mechanics, Vol. 66, Issue 4 https://doi.org/10.1007/s00466-020-01870-w	journal	August 2020
All-at-once approach to multifidelity polynomial chaos expansion surrogate modeling Bryson, Dean E.; Rumpfkeil, Markus P. Aerospace Science and Technology, Vol. 70 https://doi.org/10.1016/j.ast.2017.07.043	journal	November 2017
Multi-Index Stochastic Collocation for random PDEs Haji-Ali, Abdul-Lateef; Nobile, Fabio; Tamellini, Lorenzo Computer Methods in Applied Mechanics and Engineering, Vol. 306 https://doi.org/10.1016/j.cma.2016.03.029	journal	July 2016
Cope with diverse data structures in multi-fidelity modeling: A Gaussian process method Liu, Haitao; Ong, Yew-Soon; Cai, Jianfei Engineering Applications of Artificial Intelligence, Vol. 67 https://doi.org/10.1016/j.engappai.2017.10.008	journal	January 2018
Stochastic spectral methods for efficient Bayesian solution of inverse problems Marzouk, Youssef M.; Najm, Habib N.; Rahn, Larry A. Journal of Computational Physics, Vol. 224, Issue 2 https://doi.org/10.1016/j.jcp.2006.10.010	journal	June 2007
A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems Meng, Xuhui; Karniadakis, George Em Journal of Computational Physics, Vol. 401 https://doi.org/10.1016/j.jcp.2019.109020	journal	January 2020
A generalized approximate control variate framework for multifidelity uncertainty quantification Gorodetsky, Alex A.; Geraci, Gianluca; Eldred, Michael S. Journal of Computational Physics, Vol. 408 https://doi.org/10.1016/j.jcp.2020.109257	journal	May 2020
Predicting the output from a complex computer code when fast approximations are available Kennedy, M. Biometrika, Vol. 87, Issue 1 https://doi.org/10.1093/biomet/87.1.1	journal	March 2000
A Stochastic Collocation Algorithm with Multifidelity Models Narayan, Akil; Gittelson, Claude; Xiu, Dongbin SIAM Journal on Scientific Computing, Vol. 36, Issue 2 https://doi.org/10.1137/130929461	journal	January 2014
A Multilevel Stochastic Collocation Method for Partial Differential Equations with Random Input Data Teckentrup, A. L.; Jantsch, P.; Webster, C. G. SIAM/ASA Journal on Uncertainty Quantification, Vol. 3, Issue 1 https://doi.org/10.1137/140969002	journal	January 2015
Optimal Model Management for Multifidelity Monte Carlo Estimation Peherstorfer, Benjamin; Willcox, Karen; Gunzburger, Max SIAM Journal on Scientific Computing, Vol. 38, Issue 5 https://doi.org/10.1137/15M1046472	journal	January 2016
The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations Xiu, Dongbin; Karniadakis, George Em SIAM Journal on Scientific Computing, Vol. 24, Issue 2 https://doi.org/10.1137/S1064827501387826	journal	January 2002
Multilevel Monte Carlo Path Simulation Giles, Michael B. Operations Research, Vol. 56, Issue 3 https://doi.org/10.1287/opre.1070.0496	journal	June 2008
Recursive Co-Kriging Model for Design of Computer Experiments with Multiple Levels of Fidelity Le Gratiet, Loic; Garnier, Josselin International Journal for Uncertainty Quantification, Vol. 4, Issue 5 https://doi.org/10.1615/Int.J.UncertaintyQuantification.2014006914	journal	January 2014
MFNets: MULTI-FIDELITY DATA-DRIVEN NETWORKS FOR BAYESIAN LEARNING AND PREDICTION Gorodetsky, Alex A.; Jakeman, John D.; Geraci, Gianluca International Journal for Uncertainty Quantification, Vol. 10, Issue 6 https://doi.org/10.1615/Int.J.UncertaintyQuantification.2020032978	journal	January 2020
Multifidelity Sparse Polynomial Chaos Surrogate Models Applied to Flutter Databases Rumpfkeil, Markus P.; Beran, Philip S. AIAA Journal, Vol. 58, Issue 3 https://doi.org/10.2514/1.J058452	journal	March 2020
A multigrid approach to the optimization of systems governed by differential equations Lewis, Robert; Nash, Stephen 8th Symposium on Multidisciplinary Analysis and Optimization https://doi.org/10.2514/6.2000-4890	conference	February 2013
Second-Order Corrections for Surrogate-Based Optimization with Model Hierarchies Eldred, Michael; Giunta, Anthony; Collis, S. 10th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference https://doi.org/10.2514/6.2004-4457	conference	June 2012
Multifidelity Uncertainty Quantification Using Non-Intrusive Polynomial Chaos and Stochastic Collocation Ng, Leo Wai-Tsun; Eldred, Michael 53rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference 20th AIAA/ASME/AHS Adaptive Structures Conference 14th AIAA, 53rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference<BR>20th AIAA/ASME/AHS Adaptive Structures Conference<BR>14th AIAA https://doi.org/10.2514/6.2012-1852	conference	June 2012
Multifidelity Optimization using Statistical Surrogate Modeling for Non-Hierarchical Information Sources Lam, Rémi; Allaire, Douglas L.; Willcox, Karen E. 56th AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference https://doi.org/10.2514/6.2015-0143	conference	January 2015
A multifidelity multilevel Monte Carlo method for uncertainty propagation in aerospace applications Geraci, Gianluca; Eldred, Michael S.; Iaccarino, Gianluca 19th AIAA Non-Deterministic Approaches Conference https://doi.org/10.2514/6.2017-1951	conference	January 2017

Similar Records

MFNets: Multifidelity data-driven networks for Bayesian learning and prediction

Journal Article · 2020 · International Journal for Uncertainty Quantification · OSTI ID:1670735

Transfer learning of neural surrogates on multifidelity groundwater simulations

Dataset · 2024 · OSTI ID:3005558

Multifidelity Neural Network Formulations for Prediction of Reactive Molecular Potential Energy Surfaces

Journal Article · 2023 · Journal of Chemical Information and Modeling · OSTI ID:2311773

Related Subjects

97 MATHEMATICS AND COMPUTING
co-kriging
machine learning
multi-fidelity modeling
networks
regression
surrogate models

MFNets: data efficient all-at-once learning of multifidelity surrogates as directed networks of information sources

Citation Formats

References (27)

Similar Records

Related Subjects