Abstract
With wide applications of machine learning algorithms, machine learning security has become a significant issue. The vulnerability to adversarial perturbations exists in most machine learning algorithms, including cutting-edge deep neural networks. The standard adversarial perturbation defence techniques with adversarial training need to generate adversarial examples during the training process, which require high computational costs. This paper proposed a novel defence method using self-adaptive logit balancing and Gaussian noise boost training. This method can improve the robustness of deep neural networks without high computational cost and achieve competitive results compared with the adversarial training methods. Meanwhile, this defence method enables deep learning systems to have proactive and reactive defence during the operation. A sub-classifier is trained to determine whether the system is under attack and detect attack algorithms via the patterns of the Log-Softmax values. It can achieve high accuracy for detecting clean inputs and adversarial examples created by seven attack methods.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Aigrain, J., Detyniecki, M.: Detecting adversarial examples and other misclassifications in neural networks by introspection. arXiv preprint arXiv:1905.09186 (2019). 10.48550/arXiv. 1905.09186
Andriushchenko, M., Croce, F., Flammarion, N., Hein, M.: Square attack: a query-efficient black-box adversarial attack via random search. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12368, pp. 484–501. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58592-1_29
Athalye, A., Engstrom, L., Ilyas, A., Kwok, K.: Synthesizing robust adversarial examples. In: International conference on machine learning, pp. 284–293. PMLR (2018). https://doi.org/10.48550/arXiv.1707.07397
Carlini, N., et al.: On evaluating adversarial robustness. arXiv preprint arXiv:1902.06705 (2019). 10.48550/arXiv. 1902.06705
Carmon, Y., Raghunathan, A., Schmidt, L., Liang, P., Duchi, J.C.: Unlabeled data improves adversarial robustness. arXiv preprint arXiv:1905.13736 (2019). 10.48550/arXiv. 1905.13736
le Cessie, S., van Houwelingen, J.: Ridge estimators in logistic regression. Appl. Stat. 41(1), 191–201 (1992)
Croce, F., Hein, M.: Minimally distorted adversarial examples with a fast adaptive boundary attack. In: International Conference on Machine Learning, pp. 2196–2205. PMLR (2020). 10.48550/arXiv. 1907.02044
Croce, F., Hein, M.: Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks. In: International Conference on Machine Learning, pp. 2206–2216. PMLR (2020). 10.48550/arXiv. 2003.01690
Engstrom, L., Ilyas, A., Athalye, A.: Evaluating and understanding the robustness of adversarial logit pairing. arXiv preprint arXiv:1807.10272 (2018). 10.48550/arXiv. 1807.10272
Fu, C., Chen, H., Ruan, N., Jia, W.: Label smoothing and adversarial robustness. arXiv preprint arXiv:2009.08233 (2020). 10.48550/arXiv. 2009.08233
Goodfellow, I.J., et al.: Generative adversarial networks. arXiv preprint arXiv:1406.2661 (2014). 10.48550/arXiv. 1406.2661
Goodfellow, I.J., Shlens, J., Szegedy, C.: Explaining and harnessing adversarial examples. arXiv preprint arXiv:1412.6572 (2014). 10.48550/arXiv. 1412.6572
Hosseini, H., Kannan, S., Poovendran, R.: Are odds really odd? bypassing statistical detection of adversarial examples. arXiv preprint arXiv:1907.12138 (2019). 10.48550/arXiv. 1907.12138
Huang, X., Kwiatkowska, M., Wang, S., Wu, M.: Safety verification of deep neural networks. In: Majumdar, R., Kunčak, V. (eds.) CAV 2017. LNCS, vol. 10426, pp. 3–29. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-63387-9_1
Jin, G., Shen, S., Zhang, D., Dai, F., Zhang, Y.: APE-GAN: Adversarial perturbation elimination with gan. In: ICASSP 2019–2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 3842–3846. IEEE (2019). https://doi.org/10.1109/ICASSP.2019.8683044
Kannan, H., Kurakin, A., Goodfellow, I.: Adversarial logit pairing. arXiv preprint arXiv:1803.06373 (2018). 10.48550/arXiv. 1803.06373
Kim, H.: Torchattacks: A pytorch repository for adversarial attacks. arXiv preprint arXiv:2010.01950 (2020). 10.48550/arXiv. 2010.01950
Liu, X., Li, Y., Wu, C., Hsieh, C.J.: Adv-BNN: Improved adversarial defense through robust bayesian neural network. arXiv preprint arXiv:1810.01279 (2018). 10.48550/arXiv. 1810.01279
Madry, A., Makelov, A., Schmidt, L., Tsipras, D., Vladu, A.: Towards deep learning models resistant to adversarial attacks. arXiv preprint arXiv:1706.06083 (2017). 10.48550/arXiv. 1706.06083
Roth, K., Kilcher, Y., Hofmann, T.: The odds are odd: a statistical test for detecting adversarial examples. In: International Conference on Machine Learning, pp. 5498–5507. PMLR (2019)
Sammut, C., Webb, G.I.: Encyclopedia of Machine Learning. Springer Science & Business Media (2011)
Schmidt, L., Santurkar, S., Tsipras, D., Talwar, K., Madry, A.: Adversarially robust generalization requires more data. arXiv preprint arXiv:1804.11285 (2018). 10.48550/arXiv. 1804.11285
Shafahi, A., Ghiasi, A., Najibi, M., Huang, F., Dickerson, J.P., Goldstein, T.: Batch-wise logit-similarity: generalizing logit-squeezing and label-smoothing. In: European Conference on Computer Vision. British Machine Vision Conference (2019)
Shafahi, A., Najibi, M., Ghiasi, A., Xu, Z., Dickerson, J., Studer, C., Davis, L.S., Taylor, G., Goldstein, T.: Adversarial training for free! arXiv preprint arXiv:1904.12843 (2019). 10.48550/arXiv. 1904.12843
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., Wojna, Z.: Rethinking the inception architecture for computer vision. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2818–2826 (2016). https://doi.org/10.1109/CVPR.2016.308
Szegedy, C., et al.: Intriguing properties of neural networks. arXiv preprint arXiv:1312.6199 (2013). 10.48550/arXiv. 1312.6199
Wu, D., Xia, S.T., Wang, Y.: Adversarial weight perturbation helps robust generalization. Adv. Neural Inf. Process. Syst. 33, 2958–2969 (2020)
Zagoruyko, S., Komodakis, N.: Wide residual networks. arXiv preprint arXiv:1605.07146 (2016). 10.48550/arXiv. 1605.07146
Zhang, H., Yu, Y., Jiao, J., Xing, E., El Ghaoui, L., Jordan, M.: Theoretically principled trade-off between robustness and accuracy. In: International Conference on Machine Learning, pp. 7472–7482. PMLR (2019)
Zheng, A., Casari, A.: Feature Engineering for Machine Learning: Principles and Techniques for Data Scientists. O’Reilly Media, Inc. (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Wei, J., Meng, Q., Yao, L. (2022). Self-Adaptive Logit Balancing for Deep Learning Robustness in Computer Vision. In: Sclaroff, S., Distante, C., Leo, M., Farinella, G.M., Tombari, F. (eds) Image Analysis and Processing – ICIAP 2022. ICIAP 2022. Lecture Notes in Computer Science, vol 13231. Springer, Cham. https://doi.org/10.1007/978-3-031-06427-2_46
Download citation
DOI: https://doi.org/10.1007/978-3-031-06427-2_46
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-06426-5
Online ISBN: 978-3-031-06427-2
eBook Packages: Computer ScienceComputer Science (R0)