With the increasing deployment of deep neural networks in safety-critica...
The pretrain-finetune paradigm usually improves downstream performance o...
The high cost of acquiring and annotating samples has made the `few-shot...
Several companies often safeguard their trained deep models (i.e. detail...
Certified defense using randomized smoothing is a popular technique to
p...
Adversarial attack perturbs an image with an imperceptible noise, leadin...
Deep models are highly susceptible to adversarial attacks. Such attacks ...
Learning modality invariant features is central to the problem of
Visibl...
The utility of aerial imagery (Satellite, Drones) has become an invaluab...