Topology Adaptive Graph Convolutional Networks
This addresses a computational bottleneck in graph neural networks for researchers and practitioners, though it is incremental as it builds on existing spectral CNN methods.
The paper tackles the performance loss in spectral graph CNNs due to convolution approximations by proposing TAGCN, a vertex-domain graph convolutional network with topology-adaptive filters, achieving better performance on multiple datasets.
Spectral graph convolutional neural networks (CNNs) require approximation to the convolution to alleviate the computational complexity, resulting in performance loss. This paper proposes the topology adaptive graph convolutional network (TAGCN), a novel graph convolutional network defined in the vertex domain. We provide a systematic way to design a set of fixed-size learnable filters to perform convolutions on graphs. The topologies of these filters are adaptive to the topology of the graph when they scan the graph to perform convolution. The TAGCN not only inherits the properties of convolutions in CNN for grid-structured data, but it is also consistent with convolution as defined in graph signal processing. Since no approximation to the convolution is needed, TAGCN exhibits better performance than existing spectral CNNs on a number of data sets and is also computationally simpler than other recent methods.