TopInG: Topologically Interpretable Graph Learning via Persistent Rationale Filtration

2025-10-06

AI Summary

Graph Neural Networks (GNNs) suffer from limited interpretability in high-stakes decision-making, particularly when ground-truth reasoning relies on complex, heterogeneous rationale subgraphs—existing methods struggle to identify them accurately. To address this, we propose the first topological explanation framework grounded in persistent homology. Our method introduces adaptive topological discrepancy constraints and a reasoning-aware filtering mechanism, theoretically guaranteeing that extracted subgraphs align with the true reasoning logic. By integrating algebraic topology with autoregressive generative modeling, it robustly learns persistent topological features in graphs, effectively disentangling relevant from irrelevant substructures. Evaluated on multiple benchmark tasks, our approach maintains predictive accuracy while significantly improving explanation fidelity, enhancing detection of diverse rationale subgraphs, and mitigating spurious correlations.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Graph Neural Networks (GNNs) have shown remarkable success across various scientific fields, yet their adoption in critical decision-making is often hindered by a lack of interpretability. Recently, intrinsically interpretable GNNs have been studied to provide insights into model predictions by identifying rationale substructures in graphs. However, existing methods face challenges when the underlying rationale subgraphs are complex and varied. In this work, we propose TopInG: Topologically Interpretable Graph Learning, a novel topological framework that leverages persistent homology to identify persistent rationale subgraphs. TopInG employs a rationale filtration learning approach to model an autoregressive generation process of rationale subgraphs, and introduces a self-adjusted topological constraint, termed topological discrepancy, to enforce a persistent topological distinction between rationale subgraphs and irrelevant counterparts. We provide theoretical guarantees that our loss function is uniquely optimized by the ground truth under specific conditions. Extensive experiments demonstrate TopInG's effectiveness in tackling key challenges, such as handling variform rationale subgraphs, balancing predictive performance with interpretability, and mitigating spurious correlations. Results show that our approach improves upon state-of-the-art methods on both predictive accuracy and interpretation quality.

Problem

• Identifying persistent rationale subgraphs in complex graph data
• Balancing predictive accuracy with interpretability in graph learning
• Mitigating spurious correlations in intrinsically interpretable GNNs

Innovation

• Leverages persistent homology to identify rationale subgraphs
• Employs rationale filtration learning for autoregressive generation
• Introduces topological discrepancy constraint for persistent distinction