Abstract
Adversarial examples have become an important issue in the field of deep learning security. There have been many studies on adversarial example attack and defense algorithms for single-label classification models. However, in the real world, multi-label classification models are also widely used. There are only a few studies on adversarial example attack and defense algorithms for multi-label classification models. In this paper, we propose a negative correlation ensemble defense scheme against multi-label adversarial examples (ML-NCEn). The fundamental principle of ML-NCEn is to make the gradient directions and magnitudes of member models negatively correlated with those of other members, respectively, in the positive and negative label sets. Experimental results show that ML-NCEn has good adversarial robustness.
This study is supported by the National Key R &D Program of China (Grant No. 2022YFB3102100), Shenzhen Fundamental Research Program (Grant No. JCYJ20220818102414030), the Major Key Project of PCL (Grant No. PCL2022A03, PCL2021A02, PCL2021A09), Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies (Grant No. 2022B1212010005).
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abbasi, M., Gagné, C.: Robustness to adversarial examples through an ensemble of specialists. In: Proceedings of the 5th International Conference on Learning Representations. OpenReview (2017)
Akhtar, N., Mian, A.S.: Threat of adversarial attacks on deep learning in computer vision: a survey. IEEE Access 6, 14410–14430 (2018)
Akhtar, Z., Mouree, M.R., Dasgupta, D.: Utility of deep learning features for facial attributes manipulation detection. In: Proceedings of the IEEE International Conference on Humanized Computing and Communication with Artificial Intelligence, pp. 55–60. IEEE (2020)
Al-Qizwini, M., Barjasteh, I., Al-Qassab, H., Radha, H.: Deep Learning Algorithm for Autonomous Driving Using GoogLeNet. In: Proceedings of the IEEE Intelligent Vehicles Symposium, pp. 89–96. IEEE (2017)
Bagnall, A., Bunescu, R., Stewart, G.: Training Ensembles to Detect Adversarial Examples. CoRR abs/ arXiv: 1712.04006 (2017)
Chan, Z.S.H., Kasabov, N.K.: A preliminary study on negative correlation learning via correlation-corrected data (NCCD). Neural Process. Lett. 21(3), 207–214 (2005)
Dabouei, A., Soleymani, S., Taherkhani, F., Dawson, J.M., Nasrabadi, N.M.: Exploiting joint robustness to adversarial perturbations. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 1119–1128. IEEE, Seattle, WA, USA (2020)
Everingham, M., Eslami, S., Van Gool, L., Williams, C.K., Winn, J., Zisserman, A.: The Pascal Visual Object Classes Challenge: A Retrospective. Int. J. Comput. Vision 111(1), 98–136 (2015)
Everingham, M., Van Gool, L., Williams, C.K., Winn, J., Zisserman, A.: The Pascal Visual Object Classes (VOC) Challenge. Int. J. Comput. Vision 88(2), 303–338 (2010)
Goodfellow, I.J., Shlens, J., Szegedy, C.: Explaining and harnessing adversarial examples. In: Proceedings of the International Conference on Learning Representations, OpenReview (2015)
Kariyappa, S., Qureshi, M.K.: Improving Adversarial Robustness of Ensembles with Diversity Training. arXiv e-prints pp. arXiv-1901 (2019)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Proceedings of the Advances in Neural Information Processing Systems, pp. 1106–1114 (2012)
LeCun, Y., Bengio, Y., Hinton, G.: Deep Learning. Nature 521(7553), 436–444 (2015)
Liu, Y., Yao, X.: Ensemble learning via negative correlation. Neural Netw. 12(10), 1399–1404 (1999)
Liu, Y., Yao, X.: Simultaneous training of negatively correlated neural networks in an ensemble. IEEE Trans. Syst. Man Cybern. 29(6), 716–725 (1999)
Liu, Y., Yao, X., Higuchi, T.: Evolutionary ensembles with negative correlation learning. IEEE Trans. Evol. Comput. 4(4), 380–387 (2000)
Luo, W., Zhang, H., Kong, L., Chen, Z., Tang, K.: Defending adversarial examples by negative correlation ensemble. In: Data Mining and Big Data: 7th International Conference, DMBD 2022, Beijing, China, November 21–24 2022, Proceedings, Part II. pp. 424–438. Springer (2023). https://doi.org/10.1007/978-981-19-8991-9_30
Pang, T., Xu, K., Du, C., Chen, N., Zhu, J.: Improving adversarial robustness via promoting ensemble diversity. In: Proceedings of the 36th International Conference on Machine Learning vol. 97, pp. 4970–4979. PMLR, Long Beach, California, USA (2019)
Paszke, A., et al.: Pytorch: An imperative style, high-performance deep learning library. In: Advances in Neural Information Processing Systems 32 (2019)
Song, Q., Jin, H., Huang, X., Hu, X.: Multi-label Adversarial Perturbations. In: Proceedings of the IEEE International Conference on Data Mining, pp. 1242–1247. IEEE, Singapore (2018)
Sun, Y., Chen, Y., Wang, X., Tang, X.: Deep learning face representation by joint identification-verification. In: Proceedings of the Advances in Neural Information Processing Systems, pp. 1988–1996 (2014)
Szegedy, C., et al.: Intriguing properties of neural networks. arXiv preprint arXiv:1312.6199 (2013)
Wang, S., Chen, H., Yao, X.: Negative correlation learning for classification ensembles. In: Proceedings of the International Joint Conference on Neural Networks, pp. 1–8. IEEE, Barcelona, Spain (2010)
Yan, S., Xiong, Y., Lin, D.: Spatial temporal graph convolutional networks for skeleton-based action recognition. In: Proceedings of the AAAI Conference on Artificial Intelligence, pp. 7444–7452 (2018)
Yuan, X., He, P., Zhu, Q., Li, X.: Adversarial examples: Attacks and defenses for deep learning. IEEE Trans. Neural Netw. Learn. Syst. 30(9), 2805–2824 (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, H., Luo, W., Chen, Z., Zhou, Q. (2023). Multi-label Adversarial Defense Scheme Based on Negative Correlation Ensemble. In: Tan, Y., Shi, Y., Luo, W. (eds) Advances in Swarm Intelligence. ICSI 2023. Lecture Notes in Computer Science, vol 13969. Springer, Cham. https://doi.org/10.1007/978-3-031-36625-3_23
Download citation
DOI: https://doi.org/10.1007/978-3-031-36625-3_23
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-36624-6
Online ISBN: 978-3-031-36625-3
eBook Packages: Computer ScienceComputer Science (R0)