Skip to main content

Table 1 Comparison of Failure Prediction Model using Different Datasets

From: Ramifications of incorrect image segmentations; emphasizing on the potential effects on deep learning methods failure

Dataset AUPR-Success AUPR-Error AUC FPR-95%-TPR
MNIST
MLP
36.62 35.33 98.04 21.60
MNIST
Small ConvNet
43.26 31.03 93.44 75.63
SVHN 63.89 30.04 93.40 58.92
CIFAR-10 151.22 32.05 86.77 51.87
CIFAR-100 90.64 43.51 89.19 38.73
ConfidNet 28.36 61.54 86.84 76.21