Anomaly Detection On One Class Cifar 10

Métriques

AUROC

Résultats

Résultats de performance de divers modèles sur ce benchmark

Tableau comparatif

Nom du modèle	AUROC
gan-based-anomaly-detection-in-imbalance	90.6
explainable-deep-one-class-classification	92
unsupervised-anomaly-detection-and	91.25
generalad-anomaly-detection-across-domains-by	99.3
classification-based-anomaly-detection-for-1	88.2
anomaly-detection-requires-better	98.4
using-self-supervised-learning-can-improve	90.1
csi-novelty-detection-via-contrastive	94.3
inverse-transform-autoencoder-for-anomaly	86.6
panda-adapting-pretrained-features-for	96.2
deep-nearest-neighbor-anomaly-detection	92.5
unsupervised-anomaly-detection-and	74.33
learning-and-evaluating-representations-for-1	92.5
ssd-a-unified-framework-for-self-supervised-1	90.0
panda-adapting-pretrained-features-for	64.7
oled-one-class-learned-encoder-decoder	67.1
transformaly-two-feature-spaces-are-better	98.3
mean-shifted-contrastive-loss-for-anomaly	98.6
unsupervised-two-stage-anomaly-detection	88.4
exposing-outlier-exposure-what-can-be-learned	99.6
ocgan-one-class-novelty-detection-using-gans	66.83
fake-it-till-you-make-it-near-distribution	99.1
deep-unsupervised-image-anomaly-detection-an	92.6
panda-adapting-pretrained-features-for	64.8
p-kdgan-progressive-knowledge-distillation	73.05
when-text-and-images-don-t-mix-bias	99.1
panda-adapting-pretrained-features-for	98.9
deep-anomaly-detection-using-geometric	86
fastflow-unsupervised-anomaly-detection-and	66.7
deep-one-class-classification	65.7
esad-end-to-end-deep-semi-supervised-anomaly	83.3
anomaly-detection-via-reverse-distillation	86.5
exposing-outlier-exposure-what-can-be-learned	98.5
unsupervised-anomaly-detection-and	83.68
simple-adaptive-projection-with-pretrained	97.0
dasvdd-deep-autoencoding-support-vector-data	66.5