Out-of-Distribution Detection on STL-10

Metric: Percentage correct (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Sort:

#	Model↕	Percentage correct▼	Extra Data	Paper	Date↕	Code
1	Mixup (Gaussian)	95.93	No	On Mixup Training: Improved Calibration and Pred...	2019-05-27	Code
2	Mixup (Imagenet)	83.28	No	On Mixup Training: Improved Calibration and Pred...	2019-05-27	Code
3	Baseline (Imagenet)	80.57	No	On Mixup Training: Improved Calibration and Pred...	2019-05-27	Code
4	Dropout(Imagenet)	78.93	No	On Mixup Training: Improved Calibration and Pred...	2019-05-27	Code
5	Baseline (Gaussian)	73.28	No	On Mixup Training: Improved Calibration and Pred...	2019-05-27	Code
6	Dropout(Gaussian)	70.57	No	On Mixup Training: Improved Calibration and Pred...	2019-05-27	Code

#1Mixup (Gaussian)SOTA
95.93
Percentage correct· 2019-05-27
On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks Code
#2Mixup (Imagenet)
83.28
Percentage correct· 2019-05-27
On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks Code
#3Baseline (Imagenet)
80.57
Percentage correct· 2019-05-27
On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks Code
#4Dropout(Imagenet)
78.93
Percentage correct· 2019-05-27
On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks Code
#5Baseline (Gaussian)
73.28
Percentage correct· 2019-05-27
On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks Code
#6Dropout(Gaussian)
70.57
Percentage correct· 2019-05-27
On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks Code