Oct 6, 2021 · To predict the relation between two entities, one can use the existence of rules, namely a sequence of relations. Previous works view rules as ...

We build a novel GNN framework on the collected cycles to learn the representations of cycles, and to predict the existence/non-existence of a relation. Our ...

Inductive relation prediction is an important learning task for knowledge graph completion. To predict the relation between two entities, one can use the ...

All cycles in the input KG constitute a vector space under modulo-2 additions and multiplications. • A cycle basis is a basis spanning the cycle space.

This paper considers rules as cycles and shows that the space of cycles has a unique structure based on the mathematics of algebraic topology, ...

Jun 13, 2022 · Cycle Representation Learning for Inductive Relation Prediction good cycles. The hope is that any good cycle can be easily represented by at ...

The code for our ICML 2022 paper Cycle Representation Learning for Inductive Relation Prediction. train.py: train the model and evaluate using the AUC-PR ...

The dominant paradigm for relation prediction in knowledge graphs involves learning and operating on latent representations (i. e., embeddings) of entities and ...

The key idea of inductive relation prediction on knowledge graphs is to learn logical rules, which can capture co-occurrence patterns between relations in an ...

Cycle Representation Learning for Inductive Relation Prediction ... In this paper, we consider rules as cycles and show that the space of cycles has a unique ...