A Unified Uncertainty Representation for Graph Neural Networks via Doubly-Spectral Stochastic Expansion
DSS-GNN unifies GNN calibration, OOD detection, and shift robustness in one spectral representation.
The paper models uncertain node embeddings as random graph signals using graph Fourier filters and an orthogonal-polynomial chaos coordinate. One doubly-spectral stochastic expansion supports prediction, calibration, and an energy-based out-of-distribution score. Standalone DSS-GNN records the lowest Brier score among compared uncertainty-aware baselines on all 14 node-classification benchmarks without post-hoc correction. DSS-Hybrid achieves the best AUROC on most node-OOD settings and the strongest shifted accuracy on all seven GOOD concept-shift benchmarks under standard ERM.
- DSS expansion unifies calibration, OOD detection, and robustness in one representation.
- Standalone DSS-GNN posts the lowest Brier score on all 14 benchmarks.
- DSS-Hybrid leads shifted accuracy on all seven GOOD concept-shift benchmarks.
- A capacity theorem bounds truncation error under a full-rank feature assumption.
From the original article
1 min read at arxiv.orgReliable deployment of graph neural networks requires calibration, out-of-distribution (OOD) detection, and robustness to distribution shift, yet existing methods address these needs with separate models and objectives.
By Fred Xu · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.