Link Prediction with Relational Hypergraphs
Keywords: link prediction, relational hypergraphs, expressivity study
TL;DR: We investigate the expressive power of existing graph neural networks and propose new framework for relational hypergraph
Abstract: Link prediction with knowledge graphs has been thoroughly studied in graph machine learning, leading to a rich landscape of graph neural network architectures with successful applications.
Nonetheless, it remains challenging to transfer the success of these architectures to link prediction with *relational hypergraphs*, where the task is over *$k$-ary relations*, substantially harder than link prediction on knowledge graphs with binary relations only.
In this paper, we propose a framework for link prediction with relational hypergraphs, empowering applications of graph neural networks on *fully relational* structures. Theoretically, we conduct a thorough analysis of the expressive power of the resulting model architectures via corresponding relational Weisfeiler-Leman algorithms and also via logical expressiveness. Empirically, we validate the power of the proposed model architectures on various relational hypergraph benchmarks. The resulting model architectures substantially outperform every baseline for inductive link prediction, and also lead to state-of-the-art results for transductive link prediction.
Supplementary Material: zip
Primary Area: learning on graphs and other geometries & topologies
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 12037
Loading