Graph Alignment for Benchmarking Graph Neural Networks and Learning Positional Encodings

2025-05-19

AI Summary

This work addresses two critical challenges in graph neural network (GNN) evaluation: the absence of a universal benchmark and the low learning efficiency of positional encodings. To this end, we propose the first GNN evaluation framework that employs graph alignment as a self-supervised pretraining task. The framework supports systematic assessment of GNN architectures by generating multi-difficulty alignment datasets—both synthetic and real-world. Crucially, we formulate graph alignment as a general-purpose benchmark task and demonstrate, for the first time, its effectiveness in learning high-quality positional encodings: our method achieves state-of-the-art performance on the PCQM4Mv2 molecular property regression task with significantly fewer parameters. Moreover, anisotropic GNNs consistently outperform standard graph convolutional models on alignment tasks. To ensure reproducibility and broad adoption, we open-source a comprehensive toolkit for dataset generation and model evaluation.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Learning to Search
• Natural Language Processing: (Large) Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

We propose a novel benchmarking methodology for graph neural networks (GNNs) based on the graph alignment problem, a combinatorial optimization task that generalizes graph isomorphism by aligning two unlabeled graphs to maximize overlapping edges. We frame this problem as a self-supervised learning task and present several methods to generate graph alignment datasets using synthetic random graphs and real-world graph datasets from multiple domains. For a given graph dataset, we generate a family of graph alignment datasets with increasing difficulty, allowing us to rank the performance of various architectures. Our experiments indicate that anisotropic graph neural networks outperform standard convolutional architectures. To further demonstrate the utility of the graph alignment task, we show its effectiveness for unsupervised GNN pre-training, where the learned node embeddings outperform other positional encodings on three molecular regression tasks and achieve state-of-the-art results on the PCQM4Mv2 dataset with significantly fewer parameters. To support reproducibility and further research, we provide an open-source Python package to generate graph alignment datasets and benchmark new GNN architectures.

Problem

• Proposes benchmarking GNNs via graph alignment problem
• Generates alignment datasets for ranking GNN architectures
• Demonstrates unsupervised pre-training for molecular tasks

Innovation

• Self-supervised graph alignment for benchmarking GNNs
• Anisotropic GNNs outperform convolutional architectures
• Unsupervised pre-training improves molecular regression tasks