Adversarial Examples Detection beyond Image Space
Deep neural networks have been proved that they are vulnerable to adversarial examples, which are generated by adding human-imperceptible perturbations to images. To defend these adversarial examples, various detection based methods have been proposed. However, most of them perform poorly on detecting adversarial examples with extremely slight perturbations. By exploring these adversarial examples, we find that there exists compliance between perturbations and prediction confidence, which guides us to detect few-perturbation attacks from the aspect of prediction confidence. To detect both few-perturbation attacks and large-perturbation attacks, we propose a method beyond image space by a two-stream architecture, in which the image stream focuses on the pixel artifacts and the gradient stream copes with the confidence artifacts. The experimental results show that the proposed method outperforms the existing methods under oblivious attacks and is verified effective to defend omniscient attacks as well.
Code (1)
Similar Papers 제목 키워드 기반
Adversarial Attacks Beyond the Image Space
Generating adversarial examples is an intriguing problem and an important way of understanding the working mechanism of deep neural networks. Most existing approaches generated perturbations in the image space, i.e., eac…
Question AnsweringVisual Question AnsweringVisual Question Answering (VQA)Task-generalizable Adversarial Attack based on Perceptual Metric
Deep neural networks (DNNs) can be easily fooled by adding human imperceptible perturbations to the images. These perturbed images are known as `adversarial examples' and pose a serious threat to security and safety crit…
Adversarial Attackobject-detectionObject DetectionObject RecognitionTowards Robustness of Deep Neural Networks via Regularization
Recent studies have demonstrated the vulnerability of deep neural networks against adversarial examples. Inspired by the observation that adversarial examples often lie outside the natural image data manifold and the…
Adversarial AttackAdversarial RobustnessLearning To Characterize Adversarial Subspaces
Deep Neural Networks (DNNs) are known to be vulnerable to the maliciously generated adversarial examples. To detect these adversarial examples, previous methods use artificially designed metrics to characterize the prope…
Evading classifiers in discrete domains with provable optimality guarantees
Machine-learning models for security-critical applications such as bot, malware, or spam detection, operate in constrained discrete domains. These applications would benefit from having provable guarantees against advers…
Adversarial RobustnessSpam detectionTwitter Bot Detectionvalid