Adversarial vulnerability for any classifier
Despite achieving impressive performance, state-of-the-art classifiers remain highly vulnerable to small, imperceptible, adversarial perturbations. This vulnerability has proven empirically to be very intricate to address. In this paper, we study the phenomenon of adversarial perturbations under the assumption that the data is generated with a smooth generative model. We derive fundamental upper bounds on the robustness to perturbations of any classification function, and prove the existence of adversarial perturbations that transfer well across different classifiers with small risk. Our analysis of the robustness also provides insights onto key properties of generative models, such as their smoothness and dimensionality of latent space. We conclude with numerical experimental results showing that our bounds provide informative baselines to the maximal achievable robustness on several datasets.
Code (0)
등록된 구현이 없습니다.
Tasks
General ClassificationSimilar Papers 제목 키워드 기반
Adversarial Perturbations Are Not So Weird: Entanglement of Robust and Non-Robust Features in Neural Network Classifiers
Neural networks trained on visual data are well-known to be vulnerable to often imperceptible adversarial perturbations. The reasons for this vulnerability are still being debated in the literature. Recently Ilyas et al.…
Defending Against Adversarial Examples by Regularized Deep Embedding
Recent studies have demonstrated the vulnerability of deep convolutional neural networks against adversarial examples. Inspired by the observation that the intrinsic dimension of image data is much smaller than its pixel…
Adversarial AttackAdversarial RobustnessEnhancing Adversarial Robustness in Network Intrusion Detection: A Layer-wise Adaptive Regularization Approach
The new wave of adversarial attacks that utilize gradient-related vulnerabilities in neural network-based classifiers makes Network Intrusion Detection Systems more open to such threats. Although state-of-the-art adversa…
Network Intrusion DetectionAdversarial RobustnessAdversarial DefenseImproving Robustness of Jet Tagging Algorithms with Adversarial Training
Deep learning is a standard tool in the field of high-energy physics, facilitating considerable sensitivity enhancements for numerous analysis strategies. In particular, in identification of physics objects, such as jet …
ClassificationJet TaggingAI Classifiers Robustness to Adversarial Attacks: Manipulating Output of Classifiers by Small Input Perturbations
Deep Learning has shown an advanced performance in the state of the art applications in several domains. However, its vulnerability to adversarial attacks can remarkably ruin any powerful classifier's performance through…