Pith. sign in

REVIEW 6 cited by

Deep Semi-Supervised Anomaly Detection

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1906.02694 v2 pith:P2GBWBJM submitted 2019-06-06 cs.LG stat.ML

classification cs.LGstat.ML
keywords anomalydeepdetectionlabeledapproachesmethodsnormalsamples
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Deep approaches to anomaly detection have recently shown promising results over shallow methods on large and complex datasets. Typically anomaly detection is treated as an unsupervised learning problem. In practice however, one may have---in addition to a large set of unlabeled samples---access to a small pool of labeled samples, e.g. a subset verified by some domain expert as being normal or anomalous. Semi-supervised approaches to anomaly detection aim to utilize such labeled samples, but most proposed methods are limited to merely including labeled normal samples. Only a few methods take advantage of labeled anomalies, with existing deep approaches being domain-specific. In this work we present Deep SAD, an end-to-end deep methodology for general semi-supervised anomaly detection. We further introduce an information-theoretic framework for deep anomaly detection based on the idea that the entropy of the latent distribution for normal data should be lower than the entropy of the anomalous distribution, which can serve as a theoretical interpretation for our method. In extensive experiments on MNIST, Fashion-MNIST, and CIFAR-10, along with other anomaly detection benchmark datasets, we demonstrate that our method is on par or outperforms shallow, hybrid, and deep competitors, yielding appreciable performance improvements even when provided with only little labeled data.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Multimodal Learning for Arcing Detection in Pantograph-Catenary Systems

    cs.CV 2026-02 conditional novelty 6.0 of 10

    A multimodal image+force anomaly detector with pseudo-anomaly augmentation reports AUROCs of 93.49% on real SBB data and 95.03% on internet images plus synthetic force for pantograph-catenary arcing.

  2. Normality Calibration in Semi-supervised Graph Anomaly Detection

    cs.LG 2025-10 conditional novelty 6.0 of 10

    GraphNC calibrates normality in semi-supervised graph anomaly detection by distilling teacher anomaly scores into a student model and adding perturbation-based consistency on labeled normal nodes, outperforming prior methods.

  3. Real-Time Decorrelation-Based Anomaly Detection for Multivariate Time Series

    cs.LG 2025-07 conditional novelty 6.0 of 10

    DAD detects multivariate time-series anomalies by measuring the change in an online-learned decorrelation matrix, achieving the best mean AUC (0.8027) among 15 methods on 50 datasets.

  4. INP-Former++: Advancing Universal Anomaly Detection via Intrinsic Normal Prototypes and Residual Learning

    cs.CV 2025-06 conditional novelty 6.0 of 10

    INP-Former++ detects image defects by extracting intrinsic normal prototypes from the test image itself and reconstructing only normal regions, achieving state-of-the-art results across single-class, multi-class, few-...

  5. CAPMix: Robust KPI Anomaly Detection for AIOps in Noisy and Dynamic Environments

    cs.LG 2025-09 conditional novelty 4.0 of 10

    CAPMix combines CutAddPaste anomaly injection, DTW-based label revision, and dual-space mixup to improve time-series anomaly detection, reporting gains over prior methods on five benchmarks.

  6. Optimizing Latent Dimension Allocation in Hierarchical VAEs: Balancing Attenuation and Information Retention for OOD Detection

    cs.LG 2025-06 conditional novelty 4.0 of 10

    A dataset-specific geometric split of latent dimensions across HVAE layers improves OOD detection over fixed baseline configurations.

Pith tools