Beyond the Known: Novel Class Discovery for Open-world Graph Learning

arXiv:2403.19907v1 Announce Type: new
Abstract: Node classification on graphs is of great importance in many applications. Due to the limited labeling capability and evolution in real-world open scenarios, novel classes can emerge on unlabeled testing nodes. However, little attention has been paid to novel class discovery on graphs. Discovering novel classes is challenging as novel and known class nodes are correlated by edges, which makes their representations indistinguishable when applying message passing GNNs. Furthermore, the novel classes lack labeling information …

abstract applications arxiv attention beyond capability class classification cs.ai cs.lg discovery evolution graph graph learning graphs however importance labeling node nodes novel open-world testing type world