AlignGraph: A Group of Generative Models for Graphs

TL;DR

AlignGraph introduces a novel framework combining efficient graph alignment with permutation-invariant generative models, significantly improving the ability to learn graph distributions compared to existing methods.

Contribution

It presents a new group of generative models that effectively address permutation invariance in graph generation, outperforming previous approaches.

Findings

Successfully learns graph distributions with high accuracy.

Outperforms competitors by 25%-560% in relevant scores.

Demonstrates efficiency and effectiveness of the proposed alignment-based approach.

Abstract

It is challenging for generative models to learn a distribution over graphs because of the lack of permutation invariance: nodes may be ordered arbitrarily across graphs, and standard graph alignment is combinatorial and notoriously expensive. We propose AlignGraph, a group of generative models that combine fast and efficient graph alignment methods with a family of deep generative models that are invariant to node permutations. Our experiments demonstrate that our framework successfully learns graph distributions, outperforming competitors by 25% -560% in relevant performance scores.