Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

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

Journal Article · · Computational Mechanics

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.

Research Organization:
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA); USDOE Laboratory Directed Research and Development (LDRD) Program
Grant/Contract Number:
NA0003525
OSTI ID:
1820427
Report Number(s):
SAND-2021-10721J; 699125
Journal Information:
Computational Mechanics, Vol. 68, Issue 4; ISSN 0178-7675
Publisher:
SpringerCopyright Statement
Country of Publication:
United States
Language:
English

References (27)

A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems journal January 2020
A Stochastic Collocation Algorithm with Multifidelity Models journal January 2014
Optimal Model Management for Multifidelity Monte Carlo Estimation journal January 2016
A trust-region framework for managing the use of approximation models in optimization journal February 1998
Multifidelity Sparse Polynomial Chaos Surrogate Models Applied to Flutter Databases journal March 2020
A Multilevel Stochastic Collocation Method for Partial Differential Equations with Random Input Data journal January 2015
MFNets: MULTI-FIDELITY DATA-DRIVEN NETWORKS FOR BAYESIAN LEARNING AND PREDICTION journal January 2020
Multi-Index Stochastic Collocation for random PDEs journal July 2016
A multifidelity multilevel Monte Carlo method for uncertainty propagation in aerospace applications conference January 2017
Coupling multi-fidelity kriging and model-order reduction for the construction of virtual charts journal June 2019
The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations journal January 2002
Second-Order Corrections for Surrogate-Based Optimization with Model Hierarchies conference June 2012
A generalized approximate control variate framework for multifidelity uncertainty quantification journal May 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
Multifidelity Optimization using Statistical Surrogate Modeling for Non-Hierarchical Information Sources conference January 2015
Cope with diverse data structures in multi-fidelity modeling: A Gaussian process method journal January 2018
Multilevel Monte Carlo Path Simulation journal June 2008
Recursive Co-Kriging Model for Design of Computer Experiments with Multiple Levels of Fidelity journal January 2014
Multi-index Monte Carlo: when sparsity meets sampling journal June 2015
Stochastic spectral methods for efficient Bayesian solution of inverse problems journal June 2007
All-at-once approach to multifidelity polynomial chaos expansion surrogate modeling journal November 2017
A multigrid approach to the optimization of systems governed by differential equations conference February 2013
Predicting the output from a complex computer code when fast approximations are available journal March 2000
Bi-fidelity stochastic gradient descent for structural optimization under uncertainty journal August 2020
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
Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling journal February 2017
TIEGCM results associated with "Dynamics and electrodynamics of an UFKW packet in the ionosphere and thermosphere" dataset January 2020