Multi-SAP Adversarial Defense for Deep Neural Networks

Shorya Sharma

Abstract


Deep learning models have gained immense popularity for machine learning tasks such as image classification and natural language processing due to their high expressibility. However, they are vulnerable to adversarial samples - perturbed samples that are imperceptible to a human, but can cause the deep learning model to give incorrect predictions with a high confidence. This limitation has been a major deterrent in the deployment of deep learning algorithms in production, specifically in security critical systems. In this project, we followed a game theoretic approach to implement a novel defense strategy, that combines multiple Stochastic Activation Pruning with adversarial training. Our defense accuracy outperforms that of PGD adversarial training, which is known to be the one of the best defenses against several L∞ attacks, by about 6-7%. We are hopeful that our defense strategy can withstand strong attacks leading to more robust deep neural network models.

Deep learning models have gained immense popularity for machine learning tasks such as image classification and natural language processing due to their high expressibility. However, they are vulnerable to adversarial samples - perturbed samples that are imperceptible to a human, but can cause the deep learning model to give incorrect predictions with a high confidence. This limitation has been a major deterrent in the deployment of deep learning algorithms in production, specifically in security critical systems. In this project, we followed a game theoretic approach to implement a novel defense strategy, that combines multiple Stochastic Activation Pruning with adversarial training. Our defense accuracy outperforms that of PGD adversarial training, which is known to be the one of the best defenses against several L∞ attacks, by about 6-7%. We are hopeful that our defense strategy can withstand strong attacks leading to more robust deep neural network models.


Keywords


Adversarial attacks; adversarial defenses; stochastic activation pruning; deep neural networks; PGD; FGSM.

Full Text:

PDF

References


Ren, K., Zheng, T., Qin, Z. and Liu, X., 2020. Adversarial attacks and defenses in deep learning. Engineering.

Szegedy C, Zaremba W, Sutskever I, Bruna J, Erhan D, Goodfellow I, et al. Intriguing properties of neural networks. 2013. arXiv:1312.6199.

Goodfellow IJ, Shlens J, Szegedy C. Explaining and harnessing adversarial examples. 2014. arXiv:1412.6572.

Kurakin A, Goodfellow I, Bengio S. Adversarial examples in the physical world. 2016. arXiv:1607.02533.

Dong Y, Liao F, Pang T, Su H, Zhu J, Hu X, et al. Boosting adversarial attacks with momentum. In: Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition; 2018 Jun 18–23; Salt Lake City, UT, USA; 2018. p. 9185–193.

Carlini N, Wagner D. Towards evaluating the robustness of neural networks. In: Proceedings of the 2017 IEEE Symposium on Security and Privacy; 2017 May 22–26; San Jose, CA, USA; 2017. p. 39–57.

Zheng T, Chen C, Ren K. Distributionally adversarial attack. 2018. arXiv:1808.05537.

Moosavi-Dezfooli SM, Fawzi A, Fawzi O, Frossard P. Universal adversarial perturbations. In: Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition; 2017 Jul 21–26; Honolulu, HI, USA; 2017. p. 1765–73.

Xiao C, Li B, Zhu JY, He W, Liu M, Song D. Generating adversarial examples with adversarial networks. 2018. arXiv:1801.02610.

Sharif M, Bhagavatula S, Bauer L, Reiter MK. Accessorize to a crime: real and stealthy attacks on state-of-the-art face recognition. In: Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security; 2016 Oct 24–28; Vienna, Austria; 2016. p. 1528–40.

Parkhi OM, Vedaldi A, Zisserman A. Deep face recognition. In: Proceedings of British Machine Vision Conference; 2017 Sep 7–10; Swansea, UK; 2015.

Brown TB, Mané D, Roy A, Abadi M, Gilmer J. Adversarial patch. 2017. arXiv:1712.09665.

Madry A, Makelov A, Schmidt L, Tsipras D, Vladu A. Towards deep learning models resistant to adversarial attacks. 2017. arXiv: 1706.06083.

Athalye A, Engstrom L, Ilya A, Kwok K. Synthesizing robust adversarial examples. 2017. arXiv:1707.07397.

Dhillon GS, Azizzadenesheli K, Lipton ZC, Bernstein J, Kossaifi J, Khanna A, et al. Stochastic activation pruning for robust adversarial defense. 2018. arXiv: 1803.01442.

Xu W, Evans D, Qi Y. Feature squeezing: detecting adversarial examples in deep neural networks. 2017. arXiv: 1704.01155.

Meng D, Chen H. MagNet: a two-pronged defense against adversarial examples. In: Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security; 2017 Oct 30–Nov 3; New York, NY, USA; 2017. p. 135–47.

Neal RM. Bayesian learning for neural networks. New York: Springer Science & Business Media; 2012.

Yang Z, Li B, Chen PY, Song D. Characterizing audio adversarial examples using temporal dependency. 2018. arXiv:1809.10875.

Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. Deep Residual Learning for Image Recognition. 2015. arXiv:1512.03385

Karen Simonyan,Andrew Zisserman. Very Deep Convolutional Networks for Large Scale Image Recognition.2015.arXiv:1409.1556




DOI: https://doi.org/10.30630/ijasce.4.1.76

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Flag Counter

 

Organized / Collaboration

- Soft Computing and Data Mining Centre, UTHM, Malaysia and Department of Information Technology

- Society of Visual Informatics, Indonesia