Metraipolis › archive › dc_5028bdbebf83

How Far Can Untrained Neural Networks Go in Anomaly Detection? From Tables to Pixels — arXiv:2611.08136

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.08136
License (as archived): CC0-1.0
Archived: 2026-11-06T14:50:52.774Z
SHA-256 of the archived text: cbe38732077b25dcd8958c4865cb69df025befc051ab179a1bf1affd64f19a62
Stable URL of this page: https://metraipolis.com/documento/dc_5028bdbebf83

How to cite

How Far Can Untrained Neural Networks Go in Anomaly Detection? From Tables to Pixels — arXiv:2611.08136 — arXiv (en), https://arxiv.org/abs/2611.08136 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5028bdbebf83 (Metraipolis archive dc_5028bdbebf83, sha256 cbe38732077b25dcd8958c4865cb69df025befc051ab179a1bf1affd64f19a62)

BibTeX

@misc{dc5028bdbebf83,
  title = {How Far Can Untrained Neural Networks Go in Anomaly Detection? From Tables to Pixels — arXiv:2611.08136},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.08136},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5028bdbebf83; sha256 cbe38732077b25dcd8958c4865cb69df025befc051ab179a1bf1affd64f19a62},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08136. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Anomaly detection (AD) is often framed as a representation-learning problem, where strong performance is expected to require learned features or AD-specific optimization. However, this view can overlook the intrinsic separability of anomalies from normal references, even in the absence of learned representations. We therefore investigate how far untrained neural networks can go in anomaly detection. To this end, we propose PRISM, a framework combining fixed random nonlinear features with normal-reference Mahalanobis scoring for tabular detection and dense visual localization. Across 47 tabular datasets in ADBench, PRISM-T achieves the highest mean AUROC (86.4%) and the second-highest mean AUPRC (65.9%) among the compared methods. On MVTec AD and VisA, covering 27 visual categories, PRISM-V achieves mean image-level and pixel-level AUROCs of 90.4% and 96.4%, respectively. Notably, both models achieve these results without neural weight optimization or pretrained representations. A theoretical analysis further characterizes how nonlinear transformations affect the mean and covariance contributions to the Mahalanobis score gap, while empirical analyses examine the resulting score separability. These findings establish PRISM as both a competitive anomaly detector and a prior-to-learning baseline for assessing the added value of learned representations in anomaly detection.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2611.08136 — 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 2026-11-06T14:21:42.643Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).