This release of ART v1.8.0 introduces the first estimators for object tracking and regression, adds a general model-independent object detection estimator and new membership inference attacks.

Added

• Added estimator for object tracker GOTURN in PyTorch in art.estimators.object_tracking.PyTorchGoturn (#1318)
• Added estimator for scikit-learn DecisionTreeRegressor in art.estimators.regression.ScikitlearnDecistionTreeRegressor and added compatibility in attacks AttributeInferenceBlackBox and MembershipInferenceBlackBox (#1272)
• Added general estimator for all object detection models of torchvision in art.estimators.object_detection.PyTorchObjectDetector (#1295)
• Added membership inference attack based on boundary attacks with general threshold selection by Li and Zhang (#1197)

Changed

• Changed art.estimators.classification.BlackboxClassifier* to also accept recorded input/prediction data pairs, instead of a callable providing predictions by evaluating the attacked model, enabling attacks on prediction data only without the necessity for direct access to the attacked model (#1247)
• Moved patched Lingvo decoder to art.contrib (#1261)

Removed

• Removed art.classifiers and art.wappers, both modules have been replaced with tools in art.preprocessing.expectation_over_transformation, art.estimators.classification and art.estimators.classification.QueryEfficientGradientEstimationClassifier (#1256)

Fixed

[None]