List your company

OOD Detection on CIFAR-10 (IND) vs TinyImageNet (OOD)

94.18AUROCNGC
59.173668.261877.3586.4382Aug 25, 2021

Evaluation results

MethodLinks
NGCSupervised detection=false, IND noise level=50% sym., OOD samples in training set=20k, OOD samples in test set=10k2021.0894.18
MDSupervised detection=true, IND noise level=50% sym., OOD samples in training set=20k, OOD samples in test set=10k2021.0877.55
SSDSupervised detection=false, IND noise level=50% sym., OOD samples in training set=20k, OOD samples in test set=10k2021.0875.51
MSPSupervised detection=true, IND noise level=50% sym., OOD samples in training set=20k, OOD samples in test set=10k2021.0870.12
RotSupervised detection=false, IND noise level=50% sym., OOD samples in training set=20k, OOD samples in test set=10k2021.0868.87
ODINSupervised detection=true, IND noise level=50% sym., OOD samples in training set=20k, OOD samples in test set=10k2021.0867.31
RotSupervised detection=true, IND noise level=50% sym., OOD samples in training set=20k, OOD samples in test set=10k2021.0864.64
SSDSupervised detection=true, IND noise level=50% sym., OOD samples in training set=20k, OOD samples in test set=10k2021.0860.52