MFNets: data efficient all-at-once learning of multifidelity surrogates as directed networks of information sources
From MaRDI portal
Publication:2667288
Abstract: 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. 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.
Recommendations
- Optimal model management for multifidelity Monte Carlo estimation
- Survey of multifidelity methods in uncertainty propagation, inference, and optimization
- Surrogate modeling and multifidelity approach in computer experimentation
- Context-Aware Surrogate Modeling for Balancing Approximation and Sampling Costs in Multifidelity Importance Sampling and Bayesian Inverse Problems
- Budget-limited distribution learning in multifidelity problems
Cites work
- A composite neural network that learns from multi-fidelity data: application to function approximation and inverse PDE problems
- A generalized approximate control variate framework for multifidelity uncertainty quantification
- A multilevel stochastic collocation method for partial differential equations with random input data
- A stochastic collocation algorithm with multifidelity models
- Adaptive Multi-index Collocation for Uncertainty Quantification and Sensitivity Analysis
- An adaptive surrogate modeling based on deep neural networks for large-scale Bayesian inverse problems
- Bi-fidelity stochastic gradient descent for structural optimization under uncertainty
- Certified reduced basis methods for parametrized partial differential equations
- Coupling multi-fidelity Kriging and model-order reduction for the construction of virtual charts
- Gaussian processes for machine learning.
- MFNets: MULTI-FIDELITY DATA-DRIVEN NETWORKS FOR BAYESIAN LEARNING AND PREDICTION
- Multi-index Monte Carlo: when sparsity meets sampling
- Multi-index stochastic collocation for random PDEs
- Multilevel Monte Carlo Path Simulation
- Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling
- Optimal model management for multifidelity Monte Carlo estimation
- Predicting the output from a complex computer code when fast approximations are available
- RECURSIVE CO-KRIGING MODEL FOR DESIGN OF COMPUTER EXPERIMENTS WITH MULTIPLE LEVELS OF FIDELITY
- Statistical Results on Control Variables with Application to Queueing Network Simulation
- Stochastic spectral methods for efficient Bayesian solution of inverse problems
- The Wiener--Askey Polynomial Chaos for Stochastic Differential Equations
Cited in
(11)- Learning finite element convergence with the multi-fidelity graph neural network
- A composite neural network that learns from multi-fidelity data: application to function approximation and inverse PDE problems
- Context-aware learning of hierarchies of low-fidelity models for multi-fidelity uncertainty quantification
- Multifidelity uncertainty quantification with models based on dissimilar parameters
- Probabilistic neural data fusion for learning from an arbitrary number of multi-fidelity data sets
- Efficient multifidelity likelihood-free Bayesian inference with adaptive computational resource allocation
- Boosting efficiency and reducing graph reliance: basis adaptation integration in Bayesian multi-fidelity networks
- Multifidelity linear regression for scientific machine learning from scarce data
- The model forest ensemble Kalman filter
- Multifidelity Formulations for Triangular Transport
- Assessing the performance of correlation-based multi-fidelity neural emulators
This page was built for publication: MFNets: data efficient all-at-once learning of multifidelity surrogates as directed networks of information sources
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2667288)
