aiqbench analysis
This research introduces Neural Bayesian Anomaly Mitigation (NBAM), a novel loss function that simultaneously enhances model robustness against data contamination and provides unsupervised identification of corrupted samples. By leveraging a Bayesian latent-switch mixture model, NBAM offers calibrated per-sample posterior probabilities to distinguish clean from noisy observations without requiring explicit labels for the corruption process. The method demonstrates superior performance over standard robust losses like Huber or Student-$t$ in scenarios involving significant label contamination rates.
Neural Bayesian Anomaly Mitigation: A Robust Loss that Doubles as an Unsupervised Contamination Classifier
arXiv:2606.16524v1 Announce Type: new Abstract: Engineered robust losses such as Huber, Student-$t$, and generalised cross-entropy make supervised models tolerant of contamination but cannot answer which observations are corrupted. We introduce Neural Bayesian Anomaly Mitigation (NBAM), a general-purpose drop-in loss derived from a Bayesian latent-switch mixture model: the marginal likelihood defines a robust supervised loss, and the associated posterior defines an unsupervised contamination classifier. Like Huber or Student-$t$, NBAM can replace the standard training loss in any supervised pipeline; unlike them, it additionally learns a structured contamination model and returns a calibrated per-sample contamination posterior. A learned input-dependent prior $\pi_\phi(x)$ captures the spatial locality of contamination, so that samples near known corruptions are more likely to be flagged, while an Occam penalty emerges automatically and regularises against over-flagging. On CIFAR-10 with asymmetric label contamination, NBAM recovers the structure of the corruption process without supervision: the contamination posterior separates clean from corrupted samples, and the learned anomaly head identifies the direction of every label-flip pair. Alongside these capabilities, NBAM outperforms the four robust-loss baselines considered here at contamination rates 0.2-0.6.
Abstract:Engineered robust losses such as Huber, Student-$t$, and generalised cross-entropy make supervised models tolerant of contamination but cannot answer which observations are corrupted. We introduce Neural Bayesian Anomaly Mitigation (NBAM), a general-purpose drop-in loss derived from a Bayesian latent-switch mixture model: the marginal likelihood defines a robust supervised loss, and the associated posterior defines an unsupervised contamination classifier. Like Huber or Student-$t$, NBAM can replace the standard training loss in any supervised pipeline; unlike them, it additionally learns a structured contamination model and returns a calibrated per-sample contamination posterior. A learned input-dependent prior $\pi_\phi(x)$ captures the spatial locality of contamination, so that samples near known corruptions are more likely to be flagged, while an Occam penalty emerges automatically and regularises against over-flagging. On CIFAR-10 with asymmetric label contamination, NBAM recovers the structure of the corruption process without supervision: the contamination posterior separates clean from corrupted samples, and the learned anomaly head identifies the direction of every label-flip pair. Alongside these capabilities, NBAM outperforms the four robust-loss baselines considered here at contamination rates 0.2-0.6.
Machine Learning (cs.LG); Cosmology and Nongalactic Astrophysics (astro-ph.CO); Machine Learning (stat.ML)
arXiv:2606.16524 [cs.LG]
(or arXiv:2606.16524v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.16524 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 275 words
Related coverage
Robust Unsupervised Network Intrusion Detection via Federated Learning with Selective Aggregation under Anomalous Sample Contamination
arXiv · Jul 29, 2026
Witnesses Explain Anomalies
arXiv · Sep 5, 2026
TRACE: Artifact-Robust Statistical Shape Modeling from Imperfect Surface Scans - A Case Study in Craniosynostosis 3D Photography
arXiv · Aug 25, 2026
Decodable but Not Detectable: A Leakage Fingerprint for Near-OOD Benchmarks
arXiv · Jul 23, 2026