DeepGCNs: Making GCNs Go as Deep as CNNs
It addresses the challenge of applying deep neural networks to non-Euclidean data, which is prevalent in real-world applications like computer vision and biology, by enabling deeper GCN architectures.
This work tackled the problem of training very deep Graph Convolutional Networks (GCNs), which were limited by vanishing gradients, by transferring concepts like residual/dense connections and dilated convolutions from CNNs to GCNs, achieving promising performance with up to 112 layers in tasks such as part segmentation, semantic segmentation on point clouds, and node classification on protein-protein interaction graphs.
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.