Function Space Pooling For Graph Convolutional Networks
This addresses graph classification tasks for researchers in graph neural networks, but it is incremental as it focuses on a specific pooling technique.
The paper tackles the problem of pooling vertex representations for graph classification by proposing a novel method that maps them to a function space, and it generally outperforms most baseline methods, achieving best performance in some cases.
Convolutional layers in graph neural networks are a fundamental type of layer which output a representation or embedding of each graph vertex. The representation typically encodes information about the vertex in question and its neighbourhood. If one wishes to perform a graph centric task, such as graph classification, this set of vertex representations must be integrated or pooled to form a graph representation. In this article we propose a novel pooling method which maps a set of vertex representations to a function space representation. This method is distinct from existing pooling methods which perform a mapping to either a vector or sequence space. Experimental graph classification results demonstrate that the proposed method generally outperforms most baseline pooling methods and in some cases achieves best performance.