Metraipolis › archive › dc_3ee027feab6d
Hidden Axis of Uncertainty: Latent-Posterior Alignment in Graph Neural Networks with Bayesian Output Layers — arXiv:2608.20758
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.20758
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T10:31:04.764Z
- SHA-256 of the archived text
31d0fa00f101fdfc818615a5ef41b339c2e08820d4772f9bf2842bce89cf34da- Stable URL of this page
https://metraipolis.com/documento/dc_3ee027feab6d
How to cite
Hidden Axis of Uncertainty: Latent-Posterior Alignment in Graph Neural Networks with Bayesian Output Layers — arXiv:2608.20758 — arXiv (en), https://arxiv.org/abs/2608.20758 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3ee027feab6d (Metraipolis archive dc_3ee027feab6d, sha256 31d0fa00f101fdfc818615a5ef41b339c2e08820d4772f9bf2842bce89cf34da)
BibTeX
@misc{dc3ee027feab6d,
title = {Hidden Axis of Uncertainty: Latent-Posterior Alignment in Graph Neural Networks with Bayesian Output Layers — arXiv:2608.20758},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.20758},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3ee027feab6d; sha256 31d0fa00f101fdfc818615a5ef41b339c2e08820d4772f9bf2842bce89cf34da},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.20758. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Bayesian Neural Networks (BNNs) with Bayesian output layers provide a principled and tractable framework for quantifying predictive uncertainty, yet the mechanisms shaping that uncertainty remain unclear. While conventional theory attributes uncertainty reduction to posterior contraction, the corresponding assumptions need not hold for deep models. In the Graph Neural Networks (GNNs) with Bayesian output layers studied here, we observe that predictive uncertainty decreases as latent representations shift toward lower-variance posterior directions, even though the posterior variance does not contract. We term this behavior Latent-Posterior Alignment (LPA) and conduct interventional experiments that support its functional role in shaping predictive uncertainty. Building on this insight, we propose Alignment-Guided Learning (AGL), which explicitly promotes this alignment during training. AGL effectively reduces predictive uncertainty while preserving accuracy and improves structural calibration, ensuring that the model confidence faithfully mirrors underlying data density. These findings provide a new perspective on uncertainty dynamics in GNNs with mean-field Bayesian output layers, shifting the focus from the magnitude of the posterior to the geometric interplay between latent and parameter spaces. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-24 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.20758 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-18T10:27:55.939Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).