Re-understanding Graph Unlearning through Memorization

2026-01-21

AI Summary

This work addresses the limitations of existing graph unlearning methods, which struggle to accurately assess unlearning difficulty and employ evaluation protocols misaligned with real-world requirements. To bridge this gap, the authors introduce MGU, a memory-guided unlearning framework that leverages the memory mechanisms of graph neural networks (GNNs). MGU formulates a practical metric for unlearning difficulty by analyzing GNN memory characteristics, dynamically adjusts its optimization objective to adapt to varying task complexities, and proposes a multidimensional evaluation protocol that better reflects practical scenarios. Extensive experiments across ten real-world graph datasets demonstrate that MGU significantly outperforms state-of-the-art methods in terms of unlearning quality, computational efficiency, and preservation of model utility.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical 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 learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Graph unlearning (GU), which removes nodes, edges, or features from trained graph neural networks (GNNs), is crucial in Web applications where graph data may contain sensitive, mislabeled, or malicious information. However, existing GU methods lack a clear understanding of the key factors that determine unlearning effectiveness, leading to three fundamental limitations: (1) impractical and inaccurate GU difficulty assessment due to test-access requirements and invalid assumptions, (2) ineffectiveness on hard-to-unlearn tasks, and (3) misaligned evaluation protocols that overemphasize easy tasks and fail to capture true forgetting capability. To address these issues, we establish GNN memorization as a new perspective for understanding graph unlearning and propose MGU, a Memorization-guided Graph Unlearning framework. MGU achieves three key advances: it provides accurate and practical difficulty assessment across different GU tasks, develops an adaptive strategy that dynamically adjusts unlearning objectives based on difficulty levels, and establishes a comprehensive evaluation protocol that aligns with practical requirements. Extensive experiments on ten real-world graphs demonstrate that MGU consistently outperforms state-of-the-art baselines in forgetting quality, computational efficiency, and utility preservation.

Problem

• Graph Unlearning
• GNN Memorization
• Unlearning Difficulty
• Evaluation Protocol
• Forgetting Capability

Innovation

• Graph Unlearning
• Memorization
• GNN
• Adaptive Strategy
• Evaluation Protocol