AUTHOR=Orsini Francesco , Baracchi Daniele , Frasconi Paolo TITLE=Shift Aggregate Extract Networks JOURNAL=Frontiers in Robotics and AI VOLUME=5 YEAR=2018 URL=https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2018.00042 DOI=10.3389/frobt.2018.00042 ISSN=2296-9144 ABSTRACT=

We introduce an architecture based on deep hierarchical decompositions to learn effective representations of large graphs. Our framework extends classic R-decompositions used in kernel methods, enabling nested part-of-part relations. Unlike recursive neural networks, which unroll a template on input graphs directly, we unroll a neural network template over the decomposition hierarchy, allowing us to deal with the high degree variability that typically characterize social network graphs. Deep hierarchical decompositions are also amenable to domain compression, a technique that reduces both space and time complexity by exploiting symmetries. We show empirically that our approach is able to outperform current state-of-the-art graph classification methods on large social network datasets, while at the same time being competitive on small chemobiological benchmark datasets.