AMBER: Adaptive Mesh Generation by Iterative Mesh Resolution Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Freymuth, Niklas, Würth, Tobias, Schreiber, Nicolas, Gyenes, Balazs, Boltres, Andreas, Mitsch, Johannes, Taranovic, Aleksandar, Hoang, Tai, Dahlinger, Philipp, Becker, Philipp, Kärger, Luise, Neumann, Gerhard
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915539361726464
author Freymuth, Niklas
Würth, Tobias
Schreiber, Nicolas
Gyenes, Balazs
Boltres, Andreas
Mitsch, Johannes
Taranovic, Aleksandar
Hoang, Tai
Dahlinger, Philipp
Becker, Philipp
Kärger, Luise
Neumann, Gerhard
author_facet Freymuth, Niklas
Würth, Tobias
Schreiber, Nicolas
Gyenes, Balazs
Boltres, Andreas
Mitsch, Johannes
Taranovic, Aleksandar
Hoang, Tai
Dahlinger, Philipp
Becker, Philipp
Kärger, Luise
Neumann, Gerhard
contents The cost and accuracy of simulating complex physical systems using the Finite Element Method (FEM) scales with the resolution of the underlying mesh. Adaptive meshes improve computational efficiency by refining resolution in critical regions, but typically require task-specific heuristics or cumbersome manual design by a human expert. We propose Adaptive Meshing By Expert Reconstruction (AMBER), a supervised learning approach to mesh adaptation. Starting from a coarse mesh, AMBER iteratively predicts the sizing field, i.e., a function mapping from the geometry to the local element size of the target mesh, and uses this prediction to produce a new intermediate mesh using an out-of-the-box mesh generator. This process is enabled through a hierarchical graph neural network, and relies on data augmentation by automatically projecting expert labels onto AMBER-generated data during training. We evaluate AMBER on 2D and 3D datasets, including classical physics problems, mechanical components, and real-world industrial designs with human expert meshes. AMBER generalizes to unseen geometries and consistently outperforms multiple recent baselines, including ones using Graph and Convolutional Neural Networks, and Reinforcement Learning-based approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2505_23663
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AMBER: Adaptive Mesh Generation by Iterative Mesh Resolution Prediction
Freymuth, Niklas
Würth, Tobias
Schreiber, Nicolas
Gyenes, Balazs
Boltres, Andreas
Mitsch, Johannes
Taranovic, Aleksandar
Hoang, Tai
Dahlinger, Philipp
Becker, Philipp
Kärger, Luise
Neumann, Gerhard
Machine Learning
Computational Geometry
The cost and accuracy of simulating complex physical systems using the Finite Element Method (FEM) scales with the resolution of the underlying mesh. Adaptive meshes improve computational efficiency by refining resolution in critical regions, but typically require task-specific heuristics or cumbersome manual design by a human expert. We propose Adaptive Meshing By Expert Reconstruction (AMBER), a supervised learning approach to mesh adaptation. Starting from a coarse mesh, AMBER iteratively predicts the sizing field, i.e., a function mapping from the geometry to the local element size of the target mesh, and uses this prediction to produce a new intermediate mesh using an out-of-the-box mesh generator. This process is enabled through a hierarchical graph neural network, and relies on data augmentation by automatically projecting expert labels onto AMBER-generated data during training. We evaluate AMBER on 2D and 3D datasets, including classical physics problems, mechanical components, and real-world industrial designs with human expert meshes. AMBER generalizes to unseen geometries and consistently outperforms multiple recent baselines, including ones using Graph and Convolutional Neural Networks, and Reinforcement Learning-based approaches.
title AMBER: Adaptive Mesh Generation by Iterative Mesh Resolution Prediction
topic Machine Learning
Computational Geometry
url https://arxiv.org/abs/2505.23663