Abstract
Steganography enables a user to hide information by embedding secret messages within other non-secret texts or pictures. Recently, research along this direction has picked a new momentum when Hayes & Danezis (NIPS 2017) used adversarial learning to generate steganographic images. In adversarial learning, two neural networks are trained to learn to communicate securely in the presence of eavesdroppers (a third neural network). Hayes–Danezis forwarded this idea to steganography where two neural networks (Bob & Charlie) learn “embed” and “extract” algorithms by exchanging images with hidden text in presence of an eavesdropping neural network (Eve). Due to non-convexity of the models in the training scheme, two different machines may not learn the same embedding and extraction model even if they train on the same set of images. We take a different approach to address this issue of “robustness” in the “decryption” process. In this paper, we introduce a third neural network (Alice) who initiates the process of learning with two neural networks (Bob & Charlie). We implement and demonstrate through experiments that it is possible for Bob & Charlie to learn the same embedding and extraction model by using a new loss function and training process.
Ishak Meraouche is financially supported by the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan for his studies at Kyushu University.
Sabyasachi Dutta was financially supported by the National Institute of Information and Communications Technology (NICT), Japan, under the NICT International Invitation Program during his stay at Kyushu University where the initial phase of the research work was carried out.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Abadi, M., Andersen, D.G.: Learning to protect communications with adversarial neural cryptography. CoRR abs/1610.06918 (2016)
Coutinho, M., de Oliveira Albuquerque, R., Borges, F., García-Villalba, L.J., Kim, T.: Learning perfectly secure cryptography to protect communications with adversarial neural cryptography. Sensors 18(5), 1306 (2018)
Ghosh, A., Kulharia, V., Namboodiri, V.P., Torr, P.H., Dokania, P.K.: Multi-agent diverse generative adversarial networks. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2018)
Goodfellow, I., et al.: Generative adversarial nets. In: Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N.D., Weinberger, K.Q. (eds.) Advances in Neural Information Processing Systems, vol. 27, pp. 2672–2680. Curran Associates, Inc. (2014). http://papers.nips.cc/paper/5423-generative-adversarial-nets.pdf
Goodfellow, I.J., Shlens, J., Szegedy, C.: Explaining and harnessing adversarial examples. In: 3rd International Conference on Learning Representations, ICLR 2015, Conference Track Proceedings, San Diego, CA, USA, May 7–9, 2015 (2015)
Hayes, J., Danezis, G.: Generating steganographic images via adversarial training. In: Guyon, I., et al. (eds.) Advances in Neural Information Processing Systems, vol. 30, pp. 1954–1963. Curran Associates, Inc. (2017). http://papers.nips.cc/paper/6791-generating-steganographic-images-via-adversarial-training.pdf
Huang, C., Kairouz, P., Chen, X., Sankar, L., Rajagopal, R.: Context-aware generative adversarial privacy. CoRR abs/1710.09549 (2017). http://arxiv.org/abs/1710.09549
Kanter, I., Kinzel, W., Kanter, E.: Secure exchange of information by synchronization of neural networks. EPL (Europhys. Lett.) 57, 141 (2002)
Ke, Y., Zhang, M., Liu, J., Su, T.: Generative steganography with Kerckhoffs’ principle based on generative adversarial networks. CoRR abs/1711.04916 (2017)
Klimov, A., Mityagin, A., Shamir, A.: Analysis of neural cryptography. In: Zheng, Y. (ed.) ASIACRYPT 2002. LNCS, vol. 2501, pp. 288–298. Springer, Heidelberg (2002). https://doi.org/10.1007/3-540-36178-2_18
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. Commun. ACM 60(6), 84–90 (2017). https://doi.org/10.1145/3065386
Liu, Z., Luo, P., Wang, X., Tang, X.: Deep learning face attributes in the wild. In: Proceedings of International Conference on Computer Vision (ICCV) (December 2015)
Meraouche, I., Dutta, S., Sakurai, K.: 3-party adversarial cryptography. In: Barolli, L., Okada, Y., Amato, F. (eds.) EIDWT 2020. LNDECT, vol. 47, pp. 247–258. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-39746-3_27
Wang, X., Yang, L., Liu, R., Kadir, A.: A chaotic image encryption algorithm based on perceptron model. Nonlinear Dyn. 62, 615–621 (2010)
Yedroudj, M., Comby, F., Chaumont, M.: Steganography using a 3 player game. CoRR abs/1907.06956 (2019)
Yu, W., Cao, J.: Cryptography based on delayed chaotic neural networks. Phys. Lett. A 356(4), 333–338 (2006)
Zheng, W., Wang, K., Wang, F.: GAN-based key secret-sharing scheme in blockchain. IEEE Trans. Cybern. 1–12 (2020)
Zhou, L., Chen, J., Zhang, Y., Su, C., James, M.A.: Security analysis and new models on the intelligent symmetric key encryption. Comput. Secur. 80, 14–24 (2019)
Zhu, Y., Vargas, D.V., Sakurai, K.: Neural cryptography based on the topology evolving neural networks. In: 2018 Sixth International Symposium on Computing and Networking Workshops (CANDARW), pp. 472–478 (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Meraouche, I., Dutta, S., Sakurai, K. (2020). 3-Party Adversarial Steganography. In: You, I. (eds) Information Security Applications. WISA 2020. Lecture Notes in Computer Science(), vol 12583. Springer, Cham. https://doi.org/10.1007/978-3-030-65299-9_7
Download citation
DOI: https://doi.org/10.1007/978-3-030-65299-9_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-65298-2
Online ISBN: 978-3-030-65299-9
eBook Packages: Computer ScienceComputer Science (R0)