GESF: A Universal Discriminative Mapping Mechanism for Graph Representation Learning

TL;DR

The paper introduces GESF, a flexible graph embedding method that learns node representations without predefined neighborhood dependence, applicable to heterogeneous graphs, and demonstrates superior performance on classification benchmarks.

Contribution

GESF is a novel graph embedding approach that automatically learns the importance of neighbors and supports heterogeneous graphs, with proven theoretical guarantees.

Findings

Outperforms state-of-the-art methods on benchmark classification tasks.

Supports heterogeneous graph embedding with multiple node types.

Provides theoretical guarantees for representation capability.

Abstract

Graph embedding is a central problem in social network analysis and many other applications, aiming to learn the vector representation for each node. While most existing approaches need to specify the neighborhood and the dependence form to the neighborhood, which may significantly degrades the flexibility of representation, we propose a novel graph node embedding method (namely GESF) via the set function technique. Our method can 1) learn an arbitrary form of representation function from neighborhood, 2) automatically decide the significance of neighbors at different distances, and 3) be applied to heterogeneous graph embedding, which may contain multiple types of nodes. Theoretical guarantee for the representation capability of our method has been proved for general homogeneous and heterogeneous graphs and evaluation results on benchmark data sets show that the proposed GESF outperforms the state-of-the-art approaches on producing node vectors for classification tasks.