DeepGCNs: Making GCNs Go as Deep as CNNs

TL;DR

This paper introduces techniques to train very deep Graph Convolutional Networks (up to 112 layers) by adapting residual and dilated convolutions from CNNs, significantly improving performance on various graph-based tasks.

Contribution

It presents novel methods for training deep GCNs using residual/dense connections and dilated convolutions, enabling deeper architectures than previously possible.

Findings

Deep GCNs with up to 112 layers outperform shallow models.

Deep GCNs achieve state-of-the-art results in point cloud segmentation.

Deep GCNs improve node classification accuracy on PPI graphs.

Abstract

Convolutional Neural Networks (CNNs) have been very successful at solving a variety of computer vision tasks such as object classification and detection, semantic segmentation, activity understanding, to name just a few. One key enabling factor for their great performance has been the ability to train very deep networks. Despite their huge success in many tasks, CNNs do not work well with non-Euclidean data, which is prevalent in many real-world applications. Graph Convolutional Networks (GCNs) offer an alternative that allows for non-Eucledian data input to a neural network. While GCNs already achieve encouraging results, they are currently limited to architectures with a relatively small number of layers, primarily due to vanishing gradients during training. This work transfers concepts such as residual/dense connections and dilated convolutions from CNNs to GCNs in order to successfully train very deep GCNs. We show the benefit of using deep GCNs (with as many as 112 layers) experimentally across various datasets and tasks. Specifically, we achieve very promising performance in part segmentation and semantic segmentation on point clouds and in node classification of protein functions across biological protein-protein interaction (PPI) graphs. We believe that the insights in this work will open avenues for future research on GCNs and their application to further tasks not explored in this paper. The source code for this work is available at https://github.com/lightaime/deep_gcns_torch and https://github.com/lightaime/deep_gcns for PyTorch and TensorFlow implementation respectively.