Abstract
The vulnerabilities of deep learning models on security and privacy have attracted a lot of attentions. Researchers have revealed the possibility of reconstructing training data of a target model. However, the performances of current works are highly rely on auxiliary datasets. In this paper, we investigate the model inversion problem under a strict restriction, where the adversary aims to reconstruct plausible samples of the target class without help of auxiliary information. To solve this challenge, we propose a Gradient Sign Model Inversion (GSMI) method based on the idea of adversarial examples generation. Specifically, we make three modifications on a popular adversarial examples generation method i-FGSM to generate plausible samples. 1) increasing the number of attack iterations and 2) superposing noises to reveal more obvious features learned by target model. 3) removing subtle noises to make reconstructed samples more plausible. However, we find samples generated by GSMI still contain noisy components. Furthermore, we adopt the idea of image adjacent regions to design a two-pass components selection algorithm to generate more reasonable sample of the target class. Through experiments, we find that the inversion samples of GSMI are close to real target class samples with some fluctuations on different classes. In addition, we also provide detail analysis for reasons of limitations on the optimization-based model inversion methods.
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
Carlini, N., Wagner, D.A.: Towards evaluating the robustness of neural networks. In: 2017 IEEE Symposium on Security and Privacy, SP 2017, San Jose, CA, USA, 22–26 May 2017, pp. 39–57. IEEE Computer Society (2017)
Cotter, S.F., Kreutz-Delgado, K., Rao, B.D.: Backward sequential elimination for sparse vector subset selection. Signal Process. 81(9), 1849–1864 (2001)
Dong, Y., Pang, T., Su, H., Zhu, J.: Evading defenses to transferable adversarial examples by translation-invariant attacks. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2019, Long Beach, CA, USA, 16–20 June 2019, pp. 4312–4321. Computer Vision Foundation/IEEE (2019)
Du, M., Liu, N., Song, Q., Hu, X.: Towards explanation of DNN-based prediction with guided feature inversion. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018, London, UK, 19–23 August 2018, pp. 1358–1367. ACM (2018)
Du, Y., Fang, M., Yi, J., Cheng, J., Tao, D.: Towards query efficient black-box attacks: an input-free perspective. In: Proceedings of the 11th ACM Workshop on Artificial Intelligence and Security, CCS 2018, Toronto, ON, Canada, 19 October 2018, pp. 13–24. ACM (2018)
Eykholt, K., et al.: Robust physical-world attacks on deep learning visual classification. In: 2018 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2018, Salt Lake City, UT, USA, 18–22 June 2018, pp. 1625–1634. IEEE Computer Society (2018)
Fredrikson, M., Jha, S., Ristenpart, T.: Model inversion attacks that exploit confidence information and basic countermeasures. In: Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security, Denver, CO, USA, 12–16 October 2015, pp. 1322–1333. ACM (2015)
Fredrikson, M., Lantz, E., Jha, S., Lin, S.M., Page, D., Ristenpart, T.: Privacy in pharmacogenetics: an end-to-end case study of personalized warfarin dosing. In: Proceedings of the 23rd USENIX Security Symposium, San Diego, CA, USA, 20–22 August 2014, pp. 17–32. USENIX Association (2014)
Gilmer, J., Ford, N., Carlini, N., Cubuk, E.D.: Adversarial examples are a natural consequence of test error in noise. In: Proceedings of the 36th International Conference on Machine Learning, ICML 2019, Long Beach, California, USA, 9–15 June 2019. Proceedings of Machine Learning Research, vol. 97, pp. 2280–2289. PMLR (2019)
González, R.C., Woods, R.E.: Digital Image Processing, 3rd edn. Pearson Education, London (2008)
Goodfellow, I.J., Shlens, J., Szegedy, C.: Explaining and harnessing adversarial examples. In: 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, 7–9 May 2015, Conference Track Proceedings (2015)
Ilyas, A., Engstrom, L., Athalye, A., Lin, J.: Black-box adversarial attacks with limited queries and information. In: Proceedings of the 35th International Conference on Machine Learning, ICML 2018, Stockholmsmässan, Stockholm, Sweden, 10–15 July 2018. Proceedings of Machine Learning Research, vol. 80, pp. 2142–2151. PMLR (2018)
Kurakin, A., Goodfellow, I.J., Bengio, S.: Adversarial examples in the physical world. In: 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, 24–26 April 2017, Workshop Track Proceedings. OpenReview.net (2017)
Lecun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86(11), 2278–2324 (1998). https://doi.org/10.1109/5.726791
Mahendran, A., Vedaldi, A.: Understanding deep image representations by inverting them. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015, Boston, MA, USA, 7–12 June 2015, pp. 5188–5196. IEEE Computer Society (2015)
Papernot, N., McDaniel, P.D., Goodfellow, I.J., Jha, S., Celik, Z.B., Swami, A.: Practical black-box attacks against machine learning. In: Proceedings of the 2017 ACM on Asia Conference on Computer and Communications Security, AsiaCCS 2017, Abu Dhabi, United Arab Emirates, 2–6 April 2017, pp. 506–519. ACM (2017)
Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., Batra, D.: Grad-cam: visual explanations from deep networks via gradient-based localization. In: IEEE International Conference on Computer Vision, ICCV 2017, Venice, Italy, 22–29 October 2017, pp. 618–626. IEEE Computer Society (2017)
Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., Batra, D.: Grad-cam: visual explanations from deep networks via gradient-based localization. Int. J. Comput. Vis. 128(2), 336–359 (2020)
Szegedy, C., et al.: Intriguing properties of neural networks. In: 2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, 14–16 April 2014, Conference Track Proceedings (2014)
Wu, X., Fredrikson, M., Jha, S., Naughton, J.F.: A methodology for formalizing model-inversion attacks. In: IEEE 29th Computer Security Foundations Symposium, CSF 2016, Lisbon, Portugal, 27 June–1 July 2016, pp. 355–370. IEEE Computer Society (2016)
Xiao, H., Rasul, K., Vollgraf, R.: Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms (2017)
Yang, Z., Zhang, J., Chang, E., Liang, Z.: Neural network inversion in adversarial setting via background knowledge alignment. In: Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security, CCS 2019, London, UK, 11–15 November 2019, pp. 225–240. ACM (2019)
Zhang, Y., Jia, R., Pei, H., Wang, W., Li, B., Song, D.: The secret revealer: Generative model-inversion attacks against deep neural networks. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, 13–19 June 2020, pp. 250–258. IEEE (2020)
Zhou, B., Khosla, A., Lapedriza, À., Oliva, A., Torralba, A.: Learning deep features for discriminative localization. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, NV, USA, 27–30 June 2016, pp. 2921–2929. IEEE Computer Society (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Tian, Z., Zhang, C., Cui, L., Yu, S. (2022). GSMI: A Gradient Sign Optimization Based Model Inversion Method. In: Long, G., Yu, X., Wang, S. (eds) AI 2021: Advances in Artificial Intelligence. AI 2022. Lecture Notes in Computer Science(), vol 13151. Springer, Cham. https://doi.org/10.1007/978-3-030-97546-3_6
Download citation
DOI: https://doi.org/10.1007/978-3-030-97546-3_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-97545-6
Online ISBN: 978-3-030-97546-3
eBook Packages: Computer ScienceComputer Science (R0)