MotifNet: a motif-based Graph Convolutional Network for directed graphs
This work addresses a key bottleneck for researchers and practitioners applying graph neural networks to directed graph data, though it appears incremental as it builds on existing spectral CNN frameworks.
The authors tackled the limitation of spectral graph convolutional networks being restricted to undirected graphs by proposing MotifNet, a motif-based graph CNN for directed graphs, and demonstrated its advantage on real data.
Deep learning on graphs and in particular, graph convolutional neural networks, have recently attracted significant attention in the machine learning community. Many of such techniques explore the analogy between the graph Laplacian eigenvectors and the classical Fourier basis, allowing to formulate the convolution as a multiplication in the spectral domain. One of the key drawback of spectral CNNs is their explicit assumption of an undirected graph, leading to a symmetric Laplacian matrix with orthogonal eigendecomposition. In this work we propose MotifNet, a graph CNN capable of dealing with directed graphs by exploiting local graph motifs. We present experimental evidence showing the advantage of our approach on real data.