Efficient and Robust Classification for Sparse Attacks
Not on the map. Why?
Abstract (arXiv)
In the past two decades we have seen the popularity of neural networks increase in conjunction with their classification accuracy. Parallel to this, we have also witnessed how fragile the very same prediction models are: tiny perturbations to the inputs can cause misclassification errors throughout entire datasets. In this paper, we consider perturbations bounded by the ℓ0--norm, which have been shown as effective attacks in the domains of image-recognition, natural language processing, and malware-detection. To this end, we propose a novel defense method that consists of "truncation" and "adversarial training". We then theoretically study the Gaussian mixture setting and prove the asymptotic optimality of our proposed classifier. Motivated by the insights we obtain, we extend these components to neural network classifiers. We conduct numerical experiments in the domain of computer vision using the MNIST and CIFAR datasets, demonstrating significant improvement for the robust classification error of neural networks.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 33 of its 42 references that are in the map. First posted on arXiv 23 Jan 2022.
Builds on
33 of 42 references.
References · 33
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 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
- 2013Playing Atari with Deep Reinforcement Learningcited by 3,211
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
- 2016Adversarial examples in the physical worldcited by 1,762
- 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
- 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
- 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
- 2018Robustness May Be at Odds with Accuracycited by 873
- 2015The Limitations of Deep Learning in Adversarial Settingscited by 784
21 more
- 2017Evasion Attacks against Machine Learning at Test Timecited by 686
- 2017EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examplescited by 154
- 2018Is Robustness the Cost of Accuracy? -- A Comprehensive Study on the Robustness of 18 Deep Image Classification Modelscited by 135
- 2019Adversarial Training Can Hurt Generalizationcited by 131
- 2018Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defensescited by 109
- 2018Towards the first adversarially robust neural network model on MNISTcited by 104
- 2019Defensive Quantization: When Efficiency Meets Robustnesscited by 95
- 2019Sparse and Imperceivable Adversarial Attackscited by 95
- 2018SparseFool: a few pixels make a big differencecited by 94
- 2020Precise Tradeoffs in Adversarial Training for Linear Regressioncited by 76
- 2021Improving Adversarial Robustness via Channel-wise Activation Suppressingcited by 76
- 2019Adversarial camera stickers: A physical camera-based attack on deep learning systemscited by 69
- 2020Sparse-RS: a versatile framework for query-efficient sparse black-box adversarial attackscited by 65
- 2019Robustness Certificates for Sparse Adversarial Attacks by Randomized Ablationcited by 59
- 2019Enhancing Adversarial Defense by k-Winners-Take-Allcited by 51
- 2016Adversarial Perturbations Against Deep Neural Networks for Malware Classificationcited by 42
- 2016Simple Black-Box Adversarial Perturbations for Deep Networkscited by 42
- 2019A Simple Explanation for the Existence of Adversarial Examples with Small Hamming Distancecited by 32
- 2016Globally Normalized Transition-Based Neural Networkscited by 26
- 2018Sparsity-based Defense against Adversarial Attacks on Linear Classifierscited by 8
- 2021Robust Classification Under ℓ0 Attack for the Gaussian Mixture Modelcited by 5
Built on by
Citing papers per year: 2024: 1.
1 paper.