We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Mesh-based graph convolutional neural networks for modeling materials with microstructure

Abstract: Predicting the evolution of a representative sample of a material with microstructure is a fundamental problem in homogenization. In this work we propose a graph convolutional neural network that utilizes the discretized representation of the initial microstructure directly, without segmentation or clustering. Compared to feature-based and pixel-based convolutional neural network models, the proposed method has a number of advantages: (a) it is deep in that it does not require featurization but can benefit from it, (b) it has a simple implementation with standard convolutional filters and layers, (c) it works natively on unstructured and structured grid data without interpolation (unlike pixel-based convolutional neural networks), and (d) it preserves rotational invariance like other graph-based convolutional neural networks. We demonstrate the performance of the proposed network and compare it to traditional pixel-based convolution neural network models and feature-based graph convolutional neural networks on multiple large datasets.
Comments: 45 pages, 19 figures
Subjects: Machine Learning (cs.LG)
Cite as: arXiv:2107.00090 [cs.LG]
  (or arXiv:2107.00090v3 [cs.LG] for this version)

Submission history

From: Reese Jones [view email]
[v1] Fri, 4 Jun 2021 03:40:40 GMT (1139kb,D)
[v2] Thu, 21 Oct 2021 02:49:38 GMT (3084kb,D)
[v3] Mon, 29 Nov 2021 17:25:02 GMT (3085kb,D)

Link back to: arXiv, form interface, contact.