HypeBoy: Generative Self-Supervised Representation Learning on Hypergraphs

arXiv:2404.00638v1 Announce Type: new
Abstract: Hypergraphs are marked by complex topology, expressing higher-order interactions among multiple nodes with hyperedges, and better capturing the topology is essential for effective representation learning. Recent advances in generative self-supervised learning (SSL) suggest that hypergraph neural networks learned from generative self supervision have the potential to effectively encode the complex hypergraph topology. Designing a generative SSL strategy for hypergraphs, however, is not straightforward. Questions remain with regard to its generative SSL task, connection to downstream …

abstract advances arxiv cs.lg encode generative hypergraph interactions multiple networks neural networks nodes representation representation learning self-supervised learning ssl supervised learning supervision topology type