Improving the Long-Range Performance of Gated Graph Neural Networks
This addresses a specific bottleneck in graph neural networks for multi-relational data, but appears incremental as it builds on existing Gated Graph Neural Networks.
The paper tackled the problem of vanishing gradients in multi-relational graph neural networks, proposing a novel Gated Graph Neural Network architecture that outperforms popular models on synthetic tasks for handling long-range dependencies.
Many popular variants of graph neural networks (GNNs) that are capable of handling multi-relational graphs may suffer from vanishing gradients. In this work, we propose a novel GNN architecture based on the Gated Graph Neural Network with an improved ability to handle long-range dependencies in multi-relational graphs. An experimental analysis on different synthetic tasks demonstrates that the proposed architecture outperforms several popular GNN models.