Towards Generalised Pre-Training of Graph Models

arXiv:2311.03976v3 Announce Type: replace
Abstract: The principal benefit of unsupervised representation learning is that a pre-trained model can be fine-tuned where data or labels are scarce. Existing approaches for graph representation learning are domain specific, maintaining consistent node and edge features across the pre-training and target datasets. This has precluded transfer to multiple domains.
In this work we present Topology Only Pre-Training, a graph pre-training method based on node and edge feature exclusion. Separating graph learning into two stages, topology …

abstract arxiv benefit consistent cs.ai cs.lg data datasets domain domains edge features graph graph representation labels multiple node pre-trained model pre-training replace representation representation learning training transfer type unsupervised