You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This release of ART 1.11.0 introduces estimators for YOLO object detection and regression models, the first audio poisoning attack, new quey-efficient black-box evasion attacks, certified defenses against adversarial patch attacks, metrics quantifying membership inference and more.
Added
Added Momentum-Iterative FGSM evasion attack in MomentumIterativeMethod and added optional momentum to loss gradients in ProjectedGradientDescent* attacks. (Add MIFGSM #1614)
Added estimators for de-randomized smoothing certification against patch attacks in PyTorchDeRandomizedSmoothing and TensorFlowV2DeRandomizedSmoothing. (Derandomized smoothing #1729)
Changed check_and_transform_label_format for nb_classes=None to automatically determine the number of classes in the provided labels. (Label fixes #1747)
Added additional documentation to ZOOAttack and cleaned up the code of method compare. (Zoo documentation #1648)
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
This release of ART 1.11.0 introduces estimators for YOLO object detection and regression models, the first audio poisoning attack, new quey-efficient black-box evasion attacks, certified defenses against adversarial patch attacks, metrics quantifying membership inference and more.
Added
MomentumIterativeMethod
and added optional momentum to loss gradients inProjectedGradientDescent*
attacks. (Add MIFGSM #1614)PyTorchYolo
. (Add Estimator for YOLO object detection models in PyTorch #1715)PyTorchDeRandomizedSmoothing
andTensorFlowV2DeRandomizedSmoothing
. (Derandomized smoothing #1729)SignOPTAttack
. (Implement SIGN-OPT: A Query-Efficient Hard-label Black-box Attack #1730)SleeperAgentAttack
. (Implement sleeper agent hidden trigger backdoor attack in PyTorch #1736)ActivationDefence
. (Exclusionary Reclassification #1738)art.attacks.poisoning.perturbations.audio_perturbations
. (Dirty-Label Backdoor Poisoning Attack for Audio #1740)PyTorchRegressor
andKerasRegressor
for PyTorch and Keras. (Regression support for pytorch and keras #1651)AdversarialPatch
andAdversarialPatchNumpy
. (Add targeted option to AdversarialPatch and AdversarialPatchNumpy #1759)Changed
check_and_transform_label_format
fornb_classes=None
to automatically determine the number of classes in the provided labels. (Label fixes #1747)ZOOAttack
and cleaned up the code of methodcompare
. (Zoo documentation #1648)nb_epochs
inAdversarialTrainerMadryPGD
to match 80'000 training steps of Madry et al. (Update default value for nb_epochs in AdversarialTrainerMadryPGD #1758)Removed
[None]
Fixed
PyTorchClassifier.clone_fore_refitting
by deleting optimizer from parameters before callingset_param()
to avoid creating the cloned model with the old optimizer. (PyTorchClassifier: delete optimizer before calling set_param() #1742)nb_classes
to methodcheck_and_transform_label_format
in inference attacks. (Send nb_classes to check_and_transform_label_format in inference attacks #1713)This discussion was created from the release ART 1.11.0.
Beta Was this translation helpful? Give feedback.
All reactions