HyperAI超神経

ホームニュース最新論文チュートリアルデータセット百科事典 SOTA LLMモデル GPU ランキング学会

サイトについて

日本語

HyperAI超神経

Anomaly Detection On Unlabeled Cifar 10 Vs

評価指標

AUROC

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名	AUROC	Paper Title	Repository
Input Complexity (PixelCNN++)	53.5	Input complexity and out-of-distribution detection with likelihood-based generative models
SSD	89.6	SSD: A Unified Framework for Self-Supervised Outlier Detection
MeanShifted	90.0	Mean-Shifted Contrastive Loss for Anomaly Detection
Likelihood (Glow)	58.2	Input complexity and out-of-distribution detection with likelihood-based generative models
PsudoLabels ResNet-18	90.8	Out-of-Distribution Detection Without Class Labels	-
CSI	89.3	CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances
PsudoLabels ViT	96.7	Out-of-Distribution Detection Without Class Labels	-
PsudoLabels ResNet-152	93.3	Out-of-Distribution Detection Without Class Labels	-
Likelihood (PixelCNN++)	52.6	Input complexity and out-of-distribution detection with likelihood-based generative models
SCAN Features	90.2	Out-of-Distribution Detection Without Class Labels	-
Input Complexity (Glow)	73.6	Input complexity and out-of-distribution detection with likelihood-based generative models
GOAD	89.2	Classification-Based Anomaly Detection for General Data
MTL	82.92	Shifting Transformation Learning for Out-of-Distribution Detection	-

0 of 13 row(s) selected.