Stochastic deep collocation method based on neural architecture search and transfer learning for heterogeneous porous media

  • We present a stochastic deep collocation method (DCM) based on neural architecture search (NAS) and transfer learning for heterogeneous porous media. We first carry out a sensitivity analysis to determine the key hyper-parameters of the network to reduce the search space and subsequently employ hyper-parameter optimization to finally obtain the parameter values. The presented NAS based DCM alsoWe present a stochastic deep collocation method (DCM) based on neural architecture search (NAS) and transfer learning for heterogeneous porous media. We first carry out a sensitivity analysis to determine the key hyper-parameters of the network to reduce the search space and subsequently employ hyper-parameter optimization to finally obtain the parameter values. The presented NAS based DCM also saves the weights and biases of the most favorable architectures, which is then used in the fine-tuning process. We also employ transfer learning techniques to drastically reduce the computational cost. The presented DCM is then applied to the stochastic analysis of heterogeneous porous material. Therefore, a three dimensional stochastic flow model is built providing a benchmark to the simulation of groundwater flow in highly heterogeneous aquifers. The performance of the presented NAS based DCM is verified in different dimensions using the method of manufactured solutions. We show that it significantly outperforms finite difference methods in both accuracy and computational cost.show moreshow less

Download full text files

Export metadata

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Document Type:Article
Author:Prof. Dr.-Ing. Timon RabczukORCiDGND, Hongwei Guo, Xiaoying Zhuang, Pengwan Chen, Naif Alajlan
DOI (Cite-Link):https://doi.org/10.1007/s00366-021-01586-2Cite-Link
URN (Cite-Link):https://nbn-resolving.org/urn:nbn:de:gbv:wim2-20220209-45835Cite-Link
URL:https://link.springer.com/article/10.1007/s00366-021-01586-2
Parent Title (English):Engineering with Computers
Publisher:Springer
Place of publication:London
Language:English
Date of Publication (online):2022/02/07
Date of first Publication:2022/01/18
Release Date:2022/02/09
Publishing Institution:Bauhaus-Universität Weimar
Institutes and partner institutions:Fakultät Bauingenieurwesen / Institut für Strukturmechanik
Volume:2022
Pagenumber:26
First Page:1
Last Page:26
Tag:deep learning; neural architecture search; randomized spectral representation
GND Keyword:Maschinelles Lernen; Neuronales Lernen; Fehlerabschätzung
Dewey Decimal Classification:500 Naturwissenschaften und Mathematik
BKL-Classification:31 Mathematik / 31.80 Angewandte Mathematik
Licence (German):License Logo Creative Commons 4.0 - Namensnennung (CC BY 4.0)