Differentiable Graph Module (DGM) for Graph Convolutional Networks

Graph deep learning has recently emerged as a powerful ML concept allowing to generalize successful deep neural architectures to non-euclidean structured data. Such methods have shown promising results on a broad spectrum of applications ranging from social science, biomedicine, and particle physics to computer vision, graphics, and chemistry. One of the limitations of the majority of current graph neural network architectures is that they are often restricted to the transductive setting and rely on the assumption that the underlying graph is known and fixed. Often, this assumption is not true since the graph may be noisy, or partially and even completely unknown. In such cases, it would be helpful to infer the graph directly from the data, especially in inductive settings where some nodes were not present in the graph at training time. Furthermore, learning a graph may become an end in itself, as the inferred structure may provide complementary insights next to the downstream task. In this paper, we introduce Differentiable Graph Module (DGM), a learnable function that predicts edge probabilities in the graph which are optimal for the downstream task. DGM can be combined with convolutional graph neural network layers and trained in an end-to-end fashion. We provide an extensive evaluation of applications from the domains of healthcare (disease prediction), brain imaging (age prediction), computer graphics (3D point cloud segmentation), and computer vision (zero-shot learning). We show that our model provides a significant improvement over baselines both in transductive and inductive settings and achieves state-of-the-art results.

Automated summary

Graph deep learning is a powerful machine learning concept that generalizes deep neural architectures to non-euclidean structured data. This method has shown promising results in various applications. However, current graph neural network architectures often assume a known and fixed underlying graph, which may not be true in practice. To address this limitation, we introduce a Differentiable Graph Module (DGM) that predicts edge probabilities in the graph for optimal downstream tasks. We evaluate our model in healthcare, brain imaging, computer graphics, and computer vision domains and demonstrate its significant improvement over baselines in both transductive and inductive settings, achieving state-of-the-art results.