Skip to yearly menu bar Skip to main content


A Persistent Weisfeiler--Lehman Procedure for Graph Classification

Bastian Rieck · Christian Bock · Karsten Borgwardt

Pacific Ballroom #224

Keywords: [ Supervised Learning ] [ Others ] [ Networks and Relational Learning ] [ Kernel Methods ]


The Weisfeiler--Lehman graph kernel exhibits competitive performance in many graph classification tasks. However, its subtree features are not able to capture connected components and cycles, topological features known for characterising graphs. To extract such features, we leverage propagated node label information and transform unweighted graphs into metric ones. This permits us to augment the subtree features with topological information obtained using persistent homology, a concept from topological data analysis. Our method, which we formalise as a generalisation of Weisfeiler--Lehman subtree features, exhibits favourable classification accuracy and its improvements in predictive performance are mainly driven by including cycle information.

Live content is unavailable. Log in and register to view live content