Though the background is an important signal for image classification, o...
Many applications of reinforcement learning can be formalized as
goal-co...
In recent years, researchers have extensively studied adversarial robust...
Certified robustness in machine learning has primarily focused on advers...
Object detection plays a key role in many security-critical systems.
Adv...
The study of provable adversarial robustness for deep neural network (DN...
Randomized smoothing is a general technique for computing sample-depende...
Randomized smoothing is a popular way of providing robustness guarantees...
Randomized smoothing has been shown to provide good certified-robustness...
Adversarial training is a popular defense strategy against attack threat...
Adversarial poisoning attacks distort training data in order to corrupt ...
Patch adversarial attacks on images, in which the attacker can distort p...
Randomized smoothing, using just a simple isotropic Gaussian distributio...
Recently, techniques have been developed to provably guarantee the robus...
In the last couple of years, several adversarial attack methods based on...
Although gradient-based saliency maps are popular methods for deep learn...