arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.03569math.NAcs.LGcs.NA
收藏

GFN:一种用于多保真应用中分辨率不变降阶算子学习的图前馈网络

GFN: A graph feedforward network for resolution-invariant reduced operator learning in multifidelity applications

École Polytechnique Fédérale de Lausanne(洛桑联邦理工学院);SISSA(国际高等研究学院(SISSA))

Oisín M. Morrison, Federico Pichi, Jan S. Hesthaven

AI总结:

本文提出图前馈网络(GFN),通过将神经网络权重与网格节点直接关联,实现分辨率不变的参数化偏微分方程降阶,在多保真场景中兼具可证明误差界、轻量灵活与优异泛化性能。

摘要:

本文提出了一种新颖的面向多保真应用的分辨率不变模型降阶策略。我们的架构基于本文提出的一种新型神经网络层——图前馈网络(graph feedforward network),该网络通过在神经网络权重与网格节点之间建立直接联系,将前馈网络的概念扩展到图结构数据,从而增强了网络的可解释性。我们利用该方法能够在不同网格尺寸上进行训练和测试的能力,在基于自编码器的参数化偏微分方程降阶策略中加以应用。我们证明,这一扩展通过误差界为性能提供了可证明的保证。所提方法的能力在三个具有挑战性的基准上进行了测试,包括对流主导现象和具有高维参数空间的问题。与最先进模型相比,该方法形成了一种更轻量且高度灵活的策略,同时在单保真和多保真场景中均表现出优异的泛化性能。

英文摘要:

This work presents a novel resolution-invariant model order reduction strategy for multifidelity applications. We base our architecture on a novel neural network layer developed in this work, the graph feedforward network, which extends the concept of feedforward networks to graph-structured data by creating a direct link between the weights of a neural network and the nodes of a mesh, enhancing the interpretability of the network. We exploit the method's capability of training and testing on different mesh sizes in an autoencoder-based reduction strategy for parametrised partial differential equations. We show that this extension comes with provable guarantees on the performance via error bounds. The capabilities of the proposed methodology are tested on three challenging benchmarks, including advection-dominated phenomena and problems with a high-dimensional parameter space. The method results in a more lightweight and highly flexible strategy when compared to state-of-the-art models, while showing excellent generalisation performance in both single fidelity and multifidelity scenarios.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑