Identifying Layers Susceptible to Adversarial Attacks
Not on the map. Why?
Abstract (arXiv)
In this paper, we investigate the use of pretraining with adversarial networks, with the objective of discovering the relationship between network depth and robustness. For this purpose, we selectively retrain different portions of VGG and ResNet architectures on CIFAR-10, Imagenette, and ImageNet using non-adversarial and adversarial data. Experimental results show that susceptibility to adversarial samples is associated with low-level feature extraction layers. Therefore, retraining of high-level layers is insufficient for achieving robustness. Furthermore, adversarial attacks yield outputs from early layers that differ statistically from features for non-adversarial samples and do not permit consistent classification by subsequent layers. This supports common hypotheses regarding the association of robustness with the feature extractor, insufficiency of deeper layers in providing robustness, and large differences in adversarial and non-adversarial feature vectors.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 30 of its 36 references that are in the map. First posted on arXiv 10 Jul 2021.
Builds on
30 of 36 references.
References · 30
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
- 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
- 2014Explaining and Harnessing Adversarial Examplescited by 7,364
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
- 2013Intriguing properties of neural networkscited by 5,155
- 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,517
- 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,451
- 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,432
- 2019A Comprehensive Survey on Transfer Learningcited by 1,160
- 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 953
- 2016Adversarial Machine Learning at Scalecited by 936
18 more
- 2020Fast is better than free: Revisiting adversarial trainingcited by 727
- 2015Understanding Neural Networks Through Deep Visualizationcited by 334
- 2018Feature Denoising for Improving Adversarial Robustnesscited by 314
- 2019High Frequency Component Helps Explain the Generalization of Convolutional Neural Networkscited by 310
- 2020A Self-supervised Approach for Adversarial Robustnesscited by 180
- 2018Large Margin Deep Networks for Classificationcited by 112
- 2021On the Adversarial Robustness of Vision Transformerscited by 106
- 2019Are All Layers Created Equal?cited by 94
- 2020Smooth Adversarial Trainingcited by 89
- 2019Metric Learning for Adversarial Robustnesscited by 77
- 2021Improving Adversarial Robustness via Channel-wise Activation Suppressingcited by 76
- 2019The Odds are Odd: A Statistical Test for Detecting Adversarial Examplescited by 75
- 2019Adversarially robust transfer learningcited by 67
- 2018Excessive Invariance Causes Adversarial Vulnerabilitycited by 66
- 2017Defense against Universal Adversarial Perturbationscited by 48
- 2020A Simple Fine-tuning Is All You Need: Towards Robust Deep Learning Via Adversarial Fine-tuningcited by 30
- 2018Adversarial Defense based on Structure-to-Signal Autoencoderscited by 10
- 2020Enhancing Intrinsic Adversarial Robustness via Feature Pyramid Decodercited by 10
Built on by
Citing papers per year: 2024: 1.
1 paper.