Accueil Actualités Articles de recherche récents Tutoriels Ensembles de données Wiki SOTA Modèles LLM Classement GPU Événements

Français

Out Of Distribution Detection On Cifar 10 Vs

Métriques

AUROC

FPR95

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	AUROC	FPR95	Paper Title	Repository
ZODE-KNN	97.12	18.29	Boosting Out-of-Distribution Detection with Multiple Pre-trained Models	-
R+ViT finetuned on CIFAR-10	98.52	-	Exploring the Limits of Out-of-Distribution Detection
ViT finetuned on CIFAR-10	98.42	-	Exploring the Limits of Out-of-Distribution Detection
DHM	100	-	Deep Hybrid Models for Out-of-Distribution Detection	-
WRN 40-2 + Rotation Prediction	90.9	-	Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty
MLP-Mixer finetuned on CIFAR-10	97.85	-	Exploring the Limits of Out-of-Distribution Detection
Forte	97.63 ± 00.15	-	Forte : Finding Outliers with Representation Typicality Estimation
ERD (ResNet18)	95.1	-	Semi-supervised novelty detection using ensembles with regularized disagreement
WRN 40-2 (MSP Baseline)	87.9	-	A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
Wide 40-2 + OECC	94.9	-	Outlier Exposure with Confidence Control for Out-of-Distribution Detection
OpenHybrid	95.1	-	Hybrid Models for Open Set Recognition	-
Isolation Forest on EfficientNet Softmax values	91.95	-	Out-of-Distribution Detection Using Outlier Detection Methods
ResNet + diagonal elements of Gram matrix	79.7	-	Detecting Out-of-Distribution Examples with In-distribution Examples and Gram Matrices
WRN 40-2 + OE	93.3	-	Deep Anomaly Detection with Outlier Exposure

0 of 14 row(s) selected.