GCN-FFNN: A Two-Stream Deep Model for Learning Solution to Partial Differential Equations

TL;DR

This paper presents a novel two-stream deep learning model combining GCN and FFNN architectures to effectively learn solutions to nonlinear PDEs using both graph and grid input representations, demonstrating improved accuracy over individual models.

Contribution

Introduces a two-stream GCN-FFNN model that integrates graph and grid data for PDE solutions, with a two-phase training process for enhanced learning.

Findings

Outperforms individual GCN and FFNN models on multiple PDEs.

Effective in learning solutions inside and outside the PDE domain.

Demonstrates applicability to 1D and 2D Burgers and Schrödinger equations.

Abstract

This paper introduces a novel two-stream deep model based on graph convolutional network (GCN) architecture and feed-forward neural networks (FFNN) for learning the solution of nonlinear partial differential equations (PDEs). The model aims at incorporating both graph and grid input representations using two streams corresponding to GCN and FFNN models, respectively. Each stream layer receives and processes its own input representation. As opposed to FFNN which receives a grid-like structure, the GCN stream layer operates on graph input data where the neighborhood information is incorporated through the adjacency matrix of the graph. In this way, the proposed GCN-FFNN model learns from two types of input representations, i.e. grid and graph data, obtained via the discretization of the PDE domain. The GCN-FFNN model is trained in two phases. In the first phase, the model parameters of each stream are trained separately. Both streams employ the same error function to adjust their parameters by enforcing the models to satisfy the given PDE as well as its initial and boundary conditions on grid or graph collocation (training) data. In the second phase, the learned parameters of two-stream layers are frozen and their learned representation solutions are fed to fully connected layers whose parameters are learned using the previously used error function. The learned GCN-FFNN model is tested on test data located both inside and outside the PDE domain. The obtained numerical results demonstrate the applicability and efficiency of the proposed GCN-FFNN model over individual GCN and FFNN models on 1D-Burgers, 1D-Schr\"odinger, 2D-Burgers and 2D-Schr\"odinger equations.