Saved in:
Bibliographic Details
Main Author: Labarthe, Aldric
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.18927
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913143767171072
author Labarthe, Aldric
author_facet Labarthe, Aldric
contents Bayesian latent space models offer a principled approach to network representation, but rely on correct specification of both geometry and link function. Real-world networks often violate these assumptions, exhibiting geometric mismatch and structural anomalies that break standard metric properties. We show that such misspecification pushes the data-generating distribution outside the model class, causing Bayesian inference to become overconfident and poorly calibrated. To address this, we propose a generalized posterior framework for random geometric graphs. We introduce Link-Sequential R-SafeBayes, a method that exploits dyadic conditional independence to estimate prequential risk and adaptively tune posterior regularization. Experiments on synthetic and real-world networks demonstrate improved calibration, better link prediction performance, and a reliable criterion for selecting latent geometries across Euclidean, spherical, and hyperbolic spaces.
format Preprint
id arxiv_https___arxiv_org_abs_2605_18927
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Bayesian Latent Space Models for Graphs Are Misspecified: Toward Robust Inference via Generalized Posteriors
Labarthe, Aldric
Machine Learning
Probability
Bayesian latent space models offer a principled approach to network representation, but rely on correct specification of both geometry and link function. Real-world networks often violate these assumptions, exhibiting geometric mismatch and structural anomalies that break standard metric properties. We show that such misspecification pushes the data-generating distribution outside the model class, causing Bayesian inference to become overconfident and poorly calibrated. To address this, we propose a generalized posterior framework for random geometric graphs. We introduce Link-Sequential R-SafeBayes, a method that exploits dyadic conditional independence to estimate prequential risk and adaptively tune posterior regularization. Experiments on synthetic and real-world networks demonstrate improved calibration, better link prediction performance, and a reliable criterion for selecting latent geometries across Euclidean, spherical, and hyperbolic spaces.
title Bayesian Latent Space Models for Graphs Are Misspecified: Toward Robust Inference via Generalized Posteriors
topic Machine Learning
Probability
url https://arxiv.org/abs/2605.18927