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

被引:19
|
作者
Gorodetsky, A. A. [1 ]
Jakeman, J. D. [2 ]
Geraci, G. [2 ]
机构
[1] Univ Michigan, 3053 FXB,1320 Beal Ave, Ann Arbor, MI 48109 USA
[2] Sandia Natl Labs, Optimizat & Uncertainty Quantificat, Albuquerque, NM 87123 USA
关键词
Multi-fidelity modeling; Regression; Surrogate models; Machine learning; Networks; Co-kriging; STOCHASTIC COLLOCATION; MODEL;
D O I
10.1007/s00466-021-02042-0
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
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.
引用
收藏
页码:741 / 758
页数:18
相关论文
共 2 条