Propagation on Multi-relational Graphs for Node Regression

TL;DR

This paper introduces a novel propagation framework for node regression on multi-relational graphs, leveraging iterative neighborhood aggregation inspired by label propagation to improve the inference of missing continuous node features.

Contribution

The paper presents a new multi-relational propagation algorithm specifically designed for node regression, extending label propagation techniques to complex multi-relational, directed graphs.

Findings

Exploiting multi-relational structure improves node regression accuracy.

The proposed method outperforms baseline approaches in various scenarios.

Multi-relational data benefits from tailored propagation algorithms.

Abstract

Recent years have witnessed a rise in real-world data captured with rich structural information that can be conveniently depicted by multi-relational graphs. While inference of continuous node features across a simple graph is rather under-studied by the current relational learning research, we go one step further and focus on node regression problem on multi-relational graphs. We take inspiration from the well-known label propagation algorithm aiming at completing categorical features across a simple graph and propose a novel propagation framework for completing missing continuous features at the nodes of a multi-relational and directed graph. Our multi-relational propagation algorithm is composed of iterative neighborhood aggregations which originate from a relational local generative model. Our findings show the benefit of exploiting the multi-relational structure of the data in several node regression scenarios in different settings.