Reducing the Sensitivity of Neural Physics Simulators to Mesh Topology via Pretraining

TL;DR

This paper demonstrates that pretraining with autoencoders can significantly reduce the sensitivity of neural physics simulators to variations in mesh topology, improving their robustness across different mesh representations.

Contribution

The study introduces a pretraining approach using autoencoders with graph embeddings to mitigate mesh topology sensitivity in neural physics simulators.

Findings

Pretraining reduces sensitivity to mesh topology variations.

Autoencoder-based pretraining improves simulator robustness.

Future directions for further reducing sensitivity are discussed.

Abstract

Meshes are used to represent complex objects in high fidelity physics simulators across a variety of domains, such as radar sensing and aerodynamics. There is growing interest in using neural networks to accelerate physics simulations, and also a growing body of work on applying neural networks directly to irregular mesh data. Since multiple mesh topologies can represent the same object, mesh augmentation is typically required to handle topological variation when training neural networks. Due to the sensitivity of physics simulators to small changes in mesh shape, it is challenging to use these augmentations when training neural network-based physics simulators. In this work, we show that variations in mesh topology can significantly reduce the performance of neural network simulators. We evaluate whether pretraining can be used to address this issue, and find that employing an established autoencoder pretraining technique with graph embedding models reduces the sensitivity of neural network simulators to variations in mesh topology. Finally, we highlight future research directions that may further reduce neural simulator sensitivity to mesh topology.