Frame invariance and scalability of neural operators for partial differential equations
arXiv:2112.14769 · doi:10.4208/cicp.OA-2021-0256
Abstract
Partial differential equations (PDEs) play a dominant role in the mathematical modeling of many complex dynamical processes. Solving these PDEs often requires prohibitively high computational costs, especially when multiple evaluations must be made for different parameters or conditions. After training, neural operators can provide PDEs solutions significantly faster than traditional PDE solvers. In this work, invariance properties and computational complexity of two neural operators are examined for transport PDE of a scalar quantity. Neural operator based on graph kernel network (GKN) operates on graph-structured data to incorporate nonlocal dependencies. Here we propose a modified formulation of GKN to achieve frame invariance. Vector cloud neural network (VCNN) is an alternate neural operator with embedded frame invariance which operates on point cloud data. GKN-based neural operator demonstrates slightly better predictive performance compared to VCNN. However, GKN requires an excessively high computational cost that increases quadratically with the increasing number of discretized objects as compared to a linear increase for VCNN.
References in corpus (7)
- Graph Neural Network for Traffic Forecasting: A Survey
- Fourier Neural Operator for Parametric Partial Differential Equations
- Physics-informed graph neural Galerkin networks: A unified framework for solving PDE-governed forward and inverse problems
- Learning the solution operator of parametric partial differential equations with physics-informed DeepOnets
- Learning nonlocal constitutive models with neural networks
- Pixie: A System for Recommending 3+ Billion Items to 200+ Million Users in Real-Time
- SRINet: Learning Strictly Rotation-Invariant Representations for Point Cloud Classification and Segmentation