Table 5.

Comparison of different accuracies (%) by using different methods where cross-task adversarial (Adv) and common-corruption (CC) (Hendrycks and Dietterich, 2019) robustness transferability (RT) in CIFAR-10 → CIFAR-100 is evaluated

MethodsAdv-RTCC-RT
AARASACS-1CS-3CS-5
ACL*23.3630.8757.0052.6545.1236.46
ACL-AIR*24.2331.4457.2453.0945.9637.70
DynACL*25.0732.1356.4353.8845.7737.13
DynACL-AIR*25.6332.6258.1054.4546.9038.24
Ours (AT)26.3133.6158.4955.6247.7339.26
Ours-AIR (AT)27.6835.2259.7557.1049.0540.80

or Create an Account

Close Modal
Close Modal