Scattering GCN: Overcoming Oversmoothness in Graph Convolutional Networks
This addresses a key limitation in graph neural networks for researchers and practitioners in geometric deep learning, though it is an incremental improvement over existing approaches.
The paper tackles the problem of oversmoothing in Graph Convolutional Networks (GCNs) by augmenting them with geometric scattering transforms and residual convolutions, resulting in improved performance for semi-supervised node classification compared to leading methods like GAT.
Graph convolutional networks (GCNs) have shown promising results in processing graph data by extracting structure-aware features. This gave rise to extensive work in geometric deep learning, focusing on designing network architectures that ensure neuron activations conform to regularity patterns within the input graph. However, in most cases the graph structure is only accounted for by considering the similarity of activations between adjacent nodes, which limits the capabilities of such methods to discriminate between nodes in a graph. Here, we propose to augment conventional GCNs with geometric scattering transforms and residual convolutions. The former enables band-pass filtering of graph signals, thus alleviating the so-called oversmoothing often encountered in GCNs, while the latter is introduced to clear the resulting features of high-frequency noise. We establish the advantages of the presented Scattering GCN with both theoretical results establishing the complementary benefits of scattering and GCN features, as well as experimental results showing the benefits of our method compared to leading graph neural networks for semi-supervised node classification, including the recently proposed GAT network that typically alleviates oversmoothing using graph attention mechanisms.