Category-free Out-of-Distribution Node Detection with Feature Resonance

2025-02-22

AI Summary

Detecting out-of-distribution (OOD) nodes in graph neural networks (GNNs) without access to in-distribution (ID) class labels remains a challenging open problem. Method: This paper abandons reliance on ID class supervision and, for the first time, identifies and formalizes the “feature resonance” phenomenon—namely, that ID nodes exhibit significantly larger feature displacement during training than OOD nodes. Building upon this insight, we propose the Resonance-based Self-supervised Learning (RSL) framework, which integrates a fine-grained proxy metric for feature displacement and a synthetic OOD node co-training mechanism to enable ID-class-agnostic OOD detection. Contribution/Results: We provide theoretical analysis establishing an error bound guarantee. Extensive experiments on five real-world graph datasets demonstrate that RSL reduces false positive rate at 95% true positive rate (FPR95) by an average of 18.51%, achieving state-of-the-art performance.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Learning to Search
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

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
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Detecting out-of-distribution (OOD) nodes in the graph-based machine-learning field is challenging, particularly when in-distribution (ID) node multi-category labels are unavailable. Thus, we focus on feature space rather than label space and find that, ideally, during the optimization of known ID samples, unknown ID samples undergo more significant representation changes than OOD samples, even if the model is trained to fit random targets, which we called the Feature Resonance phenomenon. The rationale behind it is that even without gold labels, the local manifold may still exhibit smooth resonance. Based on this, we further develop a novel graph OOD framework, dubbed Resonance-based Separation and Learning (RSL), which comprises two core modules: (i) a more practical micro-level proxy of feature resonance that measures the movement of feature vectors in one training step. (ii) integrate with synthetic OOD nodes strategy to train an effective OOD classifier. Theoretically, we derive an error bound showing the superior separability of OOD nodes during the resonance period. Empirically, RSL achieves state-of-the-art performance, reducing the FPR95 metric by an average of 18.51% across five real-world datasets.

Problem

• Detecting OOD nodes without ID labels.
• Feature Resonance phenomenon in graph learning.
• RSL framework improves OOD node detection.

Innovation

• Feature Resonance phenomenon
• Resonance-based Separation and Learning
• synthetic OOD nodes strategy