Graph Attribute Aggregation Network with Progressive Margin Folding
This addresses a bottleneck in graph convolutional neural networks for researchers in graph-based machine learning, though it appears incremental as it builds on existing GCNN methods.
The paper tackles the problem of insufficient aggregation methods between graph convolution layers by introducing a Graph Attribute Aggregation Network (GAAN) with a progressive margin folding strategy, and experiments on public molecule datasets show it outperforms existing GCNN models with significant effectiveness.
Graph convolutional neural networks (GCNNs) have been attracting increasing research attention due to its great potential in inference over graph structures. However, insufficient effort has been devoted to the aggregation methods between different convolution graph layers. In this paper, we introduce a graph attribute aggregation network (GAAN) architecture. Different from the conventional pooling operations, a graph-transformation-based aggregation strategy, progressive margin folding, PMF, is proposed for integrating graph features. By distinguishing internal and margin elements, we provide an approach for implementing the folding iteratively. And a mechanism is also devised for preserving the local structures during progressively folding. In addition, a hypergraph-based representation is introduced for transferring the aggregated information between different layers. Our experiments applied to the public molecule datasets demonstrate that the proposed GAAN outperforms the existing GCNN models with significant effectiveness.