Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs
This work addresses the challenge of applying convolutional neural networks to non-Euclidean graph data, which is incremental as it builds on existing graph convolution methods.
The authors tackled the problem of generalizing convolution to arbitrary graphs by introducing edge-conditioned filters, achieving state-of-the-art results in point cloud classification and outperforming other deep learning methods on a graph classification dataset.
A number of problems can be formulated as prediction on graph-structured data. In this work, we generalize the convolution operator from regular grids to arbitrary graphs while avoiding the spectral domain, which allows us to handle graphs of varying size and connectivity. To move beyond a simple diffusion, filter weights are conditioned on the specific edge labels in the neighborhood of a vertex. Together with the proper choice of graph coarsening, we explore constructing deep neural networks for graph classification. In particular, we demonstrate the generality of our formulation in point cloud classification, where we set the new state of the art, and on a graph classification dataset, where we outperform other deep learning approaches. The source code is available at https://github.com/mys007/ecc