Download PDFOpen PDF in browser

An OOD Multi-Task Perspective for Link Prediction with New Relation Types and Nodes

EasyChair Preprint 10556

23 pagesDate: July 13, 2023

Abstract

The task of inductive link prediction in (discrete) attributed multigraphs infers missing attributed links (relations) between nodes in new test multigraphs. Traditional relational learning methods face the challenge of limited generalization to OOD test multigraphs containing both novel nodes and novel relation types not seen in training. Recently, under the only assumption that all relation types share the same structural predictive patterns (single task), Gao et al. (2023) proposed an OOD link prediction method using the theoretical concept of double exchangeability (for nodes & relation types), in contrast to the (single) exchangeability (only for nodes) used to design Graph Neural Networks (GNNs). In this work we further extend the double exchangeability concept to multi-task double exchangeability, where we define link prediction in attributed multigraphs that can have distinct and potentially conflicting predictive patterns for different sets of relation types (multiple tasks). Our empirical results on real-world datasets demonstrate that our approach can effectively generalize to entirely new relation types in test, without access to additional information, yielding significant performance improvements over existing methods.

Keyphrases: Equivariance, Knowledge Graphs, exchangeability

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@booklet{EasyChair:10556,
  author    = {Jincheng Zhou and Beatrice Bevilacqua and Bruno Ribeiro},
  title     = {An OOD Multi-Task Perspective for Link Prediction with New Relation Types and Nodes},
  howpublished = {EasyChair Preprint 10556},
  year      = {EasyChair, 2023}}
Download PDFOpen PDF in browser