Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection
This addresses a practical problem in graph machine learning for applications requiring OOD detection without labeled data, representing a novel methodological contribution rather than an incremental improvement.
The paper tackles the challenge of detecting out-of-distribution (OOD) nodes in graph-based machine learning when in-distribution labels are unavailable, by introducing the Feature Resonance phenomenon and a framework called RSL that achieves state-of-the-art performance across thirteen real-world graph datasets.
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. Extensive experiments on a total of thirteen real-world graph datasets empirically demonstrate that RSL achieves state-of-the-art performance.