SOTAVerified

Unsupervised Anomaly Detection

The objective of Unsupervised Anomaly Detection is to detect previously unseen rare objects or events without any prior knowledge about these. The only information available is that the percentage of anomalies in the dataset is small, usually less than 1%. Since anomalies are rare and unknown to the user at training time, anomaly detection in most cases boils down to the problem of modelling the normal data distribution and defining a measurement in this space in order to classify samples as anomalous or normal. In high-dimensional data such as images, distances in the original space quickly lose descriptive power (curse of dimensionality) and a mapping to some more suitable space is required.

Source: Unsupervised Learning of Anomaly Detection from Contaminated Image Data using Simultaneous Encoder Training

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 251–275 of 506 papers

TitleStatusHype
Knowledge Distillation for Anomaly Detection—0
LARA: A Light and Anti-overfitting Retraining Approach for Unsupervised Time Series Anomaly Detection—0
Excision And Recovery: Visual Defect Obfuscation Based Self-Supervised Anomaly Detection Strategy—0
(Predictable) Performance Bias in Unsupervised Anomaly Detection—0
STemGAN: spatio-temporal generative adversarial network for video anomaly detection—0
Resilient VAE: Unsupervised Anomaly Detection at the SLAC Linac Coherent Light Source—0
Towards frugal unsupervised detection of subtle abnormalities in medical imagingCode0
Deep Semi-Supervised Anomaly Detection for Finding Fraud in the Futures Market—0
Neural Network Training Strategy to Enhance Anomaly Detection Performance: A Perspective on Reconstruction Loss AmplificationCode0
Self-Supervision for Tackling Unsupervised Anomaly Detection: Pitfalls and Opportunities—0
Bias in Unsupervised Anomaly Detection in Brain MRI—0
A Generic Machine Learning Framework for Fully-Unsupervised Anomaly Detection with Contaminated Data—0
Contaminated Multivariate Time-Series Anomaly Detection with Spatio-Temporal Graph Conditional Diffusion Models—0
Semi-Supervised Anomaly Detection for the Determination of Vehicle Hijacking Tweets—0
Forensic Data Analytics for Anomaly Detection in Evolving Networks—0
A Graph Encoder-Decoder Network for Unsupervised Anomaly Detection—0
Achieving state-of-the-art performance in the Medical Out-of-Distribution (MOOD) challenge using plausible synthetic anomaliesCode0
Patch-wise Auto-Encoder for Visual Anomaly Detection—0
AMAE: Adaptation of Pre-Trained Masked Autoencoder for Dual-Distribution Anomaly Detection in Chest X-Rays—0
TabADM: Unsupervised Tabular Anomaly Detection with Diffusion Models—0
LafitE: Latent Diffusion Model with Feature Editing for Unsupervised Multi-class Anomaly Detection—0
DSV: An Alignment Validation Loss for Self-supervised Outlier Model SelectionCode0
Unsupervised Learning of Distributional Properties can Supplement Human Labeling and Increase Active Learning Efficiency in Anomaly Detection—0
Personalized Anomaly Detection in PPG Data using Representation Learning and Biometric Identification—0
Exploring Dual Model Knowledge Distillation for Anomaly Detection—0
Show:102550
← PrevPage 11 of 21Next →
All datasetsAnoShiftSMAPVehicle ClaimsFashion-MNISTKolektorSDD220NEWSAeBAD-SCaltech-101DAGM2007ECG5000KolektorSDDMNIST

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ACR-NTL (zero-shot, test anomaly ratio=1%)ROC-AUC FAR62.5—Unverified
2ACR-DSVDD (zero-shot, anomaly ratio=1%)ROC-AUC FAR62—Unverified
3ACR-NTL (zero-shot, test anomaly ratio=20%)ROC-AUC FAR62—Unverified
4ACR-DSVDD (zero-shot, anomaly ratio=20%)ROC-AUC FAR59.1—Unverified
5COPODROC-AUC FAR50.42—Unverified
6OC-SVMROC-AUC FAR49.57—Unverified
7SO-GAALROC-AUC FAR49.35—Unverified
8ECOD Li et al. (2022)ROC-AUC FAR49.19—Unverified
9LOFROC-AUC FAR34.96—Unverified
10deepSVDDROC-AUC FAR34.53—Unverified
#ModelMetricClaimedVerifiedStatus
1DFM (flow matching)F194.1—Unverified
2ContextFlow++ (Glow-based)F193.62—Unverified
3TranAdF189.15—Unverified
4MTAD-GATF188.8—Unverified
5CAE-MF188.27—Unverified
6OmniAnomalyF187.28—Unverified
7GlowF186.05—Unverified
8GDNF185.18—Unverified
9USADF181.86—Unverified
#ModelMetricClaimedVerifiedStatus
1SOMAUC65.43—Unverified
2Isolation ForestAUC59.42—Unverified
3Latent Outlier ExposureAUC58.59—Unverified
4NeuTraL-ADAUC57.03—Unverified
5RSRAEAUC55.38—Unverified
6SOM-DAGMMAUC53.82—Unverified
7Local Outlier FactorAUC52.86—Unverified
8One Class Support Vector MachinesAUC51.68—Unverified
9DAGMMAUC51.22—Unverified
#ModelMetricClaimedVerifiedStatus
1RSRAEAUC-ROC0.85—Unverified
2RSRAEAUC (outlier ratio = 0.5)0.83—Unverified
3RSRAEAUC-ROC0.75—Unverified
4RSRAEAUC-ROC0.69—Unverified
5RSRAEAUC-ROC0.69—Unverified
#ModelMetricClaimedVerifiedStatus
1Semi-orthogonalSegmentation AUROC98.1—Unverified
2WeakREST-UnSegmentation AP76.9—Unverified
3DSRSegmentation AP61.4—Unverified
#ModelMetricClaimedVerifiedStatus
1RSRAEAUC (outlier ratio = 0.5)0.83—Unverified
#ModelMetricClaimedVerifiedStatus
1MSFRDetection AUROC87.1—Unverified
#ModelMetricClaimedVerifiedStatus
1RSRAEAUC (outlier ratio = 0.5)0.77—Unverified
#ModelMetricClaimedVerifiedStatus
1DiffusionADDetection AUROC99.6—Unverified
#ModelMetricClaimedVerifiedStatus
1VRAE+SVMAUC0.98—Unverified
#ModelMetricClaimedVerifiedStatus
1Semi-orthogonalSegmentation AUROC96—Unverified
#ModelMetricClaimedVerifiedStatus
1LVADAUROC0.94—Unverified
#ModelMetricClaimedVerifiedStatus
1DyEdgeGATAUC0.8—Unverified
#ModelMetricClaimedVerifiedStatus
1RSRAEAUC (outlier ratio = 0.5)0.85—Unverified
#ModelMetricClaimedVerifiedStatus
1TranADPrecision92.62—Unverified
#ModelMetricClaimedVerifiedStatus
1LVADAUC-ROC1—Unverified
#ModelMetricClaimedVerifiedStatus
1DyEdgeGATAUC0.83—Unverified
#ModelMetricClaimedVerifiedStatus
1P-CAE W-MSE (Tilted View)AUROC78.1—Unverified