Utilizing Adversarial Targeted Attacks to Boost Adversarial Robustness
Not on the map. Why?
Abstract (arXiv)
Adversarial attacks have been shown to be highly effective at degrading the performance of deep neural networks (DNNs). The most prominent defense is adversarial training, a method for learning a robust model. Nevertheless, adversarial training does not make DNNs immune to adversarial perturbations. We propose a novel solution by adopting the recently suggested Predictive Normalized Maximum Likelihood. Specifically, our defense performs adversarial targeted attacks according to different hypotheses, where each hypothesis assumes a specific label for the test sample. Then, by comparing the hypothesis probabilities, we predict the label. Our refinement process corresponds to recent findings of the adversarial subspace properties. We extensively evaluate our approach on 16 adversarial attack benchmarks using ResNet-50, WideResNet-28, and a2-layer ConvNet trained with ImageNet, CIFAR10, and MNIST, showing a significant improvement of up to 5.7%, 3.7%, and 0.6% respectively.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 24 of its 32 references that are in the map. First posted on arXiv 4 Sep 2021.
Builds on
24 of 32 references.
References · 24
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2013Intriguing properties of neural networkscited by 5,114
- 2016Wide Residual Networkscited by 3,057
- 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
- 2016Practical Black-Box Attacks against Machine Learningcited by 893
- 2020Fast is better than free: Revisiting adversarial trainingcited by 723
- 2017Evasion Attacks against Machine Learning at Test Timecited by 686
- 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
- 2017Countering Adversarial Images using Input Transformationscited by 517
- 2019On Evaluating Adversarial Robustnesscited by 442
12 more
- 2019Unlabeled Data Improves Adversarial Robustnesscited by 415
- 2018Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Modelscited by 362
- 2019HopSkipJumpAttack: A Query-Efficient Decision-Based Attackcited by 328
- 2017PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examplescited by 243
- 2019Adversarial Robustness through Local Linearizationcited by 146
- 2019Defense Against Adversarial Attacks Using Feature Scattering-based Adversarial Trainingcited by 95
- 2018Improving DNN Robustness to Adversarial Attacks using Jacobian Regularizationcited by 88
- 2015Exploring the Space of Adversarial Imagescited by 28
- 2019Deep pNML: Predictive Normalized Maximum Likelihood for Deep Neural Networkscited by 12
- 2019A New Look at an Old Problem: A Universal Learning Approach to Linear Regressioncited by 11
- 2019Balancing Specialization, Generalization, and Compression for Detection and Trackingcited by 6
- 2021Learning Rotation Invariant Features for Cryogenic Electron Microscopy Image Reconstructioncited by 5
Built on by
Citing papers per year: 2021: 1; 2022: 1.
2 papers.