MOTIF-Driven Contrastive Learning of Graph Representations
Arjun Subramonian
Abstract
We propose a MOTIF-driven contrastive framework to pretrain a graph neural network in a self-supervised manner so that it can automatically mine motifs from large graph datasets. Our framework achieves state-of-the-art results on various graph-level downstream tasks with few labels, like molecular property prediction.
Topics & Concepts
Motif (music)Computer scienceGraphArtificial intelligenceNatural language processingTheoretical computer sciencePhysicsAcousticsAdvanced Graph Neural NetworksGraph Theory and AlgorithmsComplex Network Analysis Techniques