Abstract
In CRYPTO 2019, Gohr shows that well-trained neural networks can perform cryptanalytic distinguishing tasks superior to traditional differential distinguishers. Moreover, applying an unorthodox key guessing strategy, an 11-round key-recovery attack on a modern block cipher Speck32/64 improves upon the published state-of-the-art result. This calls into the next questions. To what extent is the advantage of machine learning (ML) over traditional methods, and whether the advantage generally exists in the cryptanalysis of modern ciphers? To answer the first question, we devised ML-based key-recovery attacks on more extended round-reduced Speck32/64. We achieved an improved 12-round and the first practical 13-round attacks. The essential for the new results is enhancing a classical component in the ML-based attacks, that is, the neutral bits. To answer the second question, we produced various neural distinguishers on round-reduced Simon32/64 and provided comparisons with their pure differential-based counterparts.
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
Notes
- 1.
Speck can be represented as a composition of two Feistel maps [6].
- 2.
For Speck, there is no whitening key and the first subkey is XORed after the first non-linear operation, which makes the first round free in differential attack (see the top of Fig. 3 in [5]).
- 3.
In these considered variables, \(({x_1}, {y_1}) = (\tilde{x}_1\oplus k_0, \tilde{y}_1\oplus k_0)\) is the real input to the \(\mathcal{C}\mathcal{D}\) (see Fig. 3 in [5]), where \((\tilde{x}_1, \tilde{y}_1)\) is the chosen data in the key-recovery attack (since, in the key-recovery attack, the \(\mathcal{C}\mathcal{D}\) will be freely extended one round backward).
- 4.
Since the first two 3-round \(\mathcal{C}\mathcal{D}\)s are used as paired differentials, the key bit \(k_0[5] \oplus k_0[14]\) does not need to be guessed. Besides, since the CSNBS [6, 11, 12, 18] in Table 4 is not used in the attack, the key bit \(k_0[2] \oplus k_0[11]\) does not need to be guessed. In total only 5 bits of \(k_0\) are guessed.
- 5.
Tesla V100-SXM2-32GB, computeCapability: 7.0; coreClock: 1.53 GHz; coreCount: 80; deviceMemorySize: 31.72 GB; deviceMemoryBandwidth: 836.37 GB/s).
- 6.
Under the assumption that one second equals the time of \(2^{28}\) executions of Speck32/64 on a CPU, and \(r = \log _{2}(cpu/gpu)\), where cpu is the CPU time and gpu is the GPU time running an attack. In our computing systems, \(r=2.4\).
- 7.
Equipped with a 32-core Intel Cascade-Lake Xeon(R) Platinum 9221 2.30 GHz, and with 384 GB RAM, on CentOS 7.6.
- 8.
Some \({v_2}_{\textrm{max}}\)’s corresponding to success cases are lower than cutoff \(c_2\); that is due to the final improvement.
References
Abadi, M., Andersen, D.G.: Learning to protect communications with adversarial neural cryptography. arXiv preprint arXiv:1610.06918 (2016)
Abed, F., List, E., Lucks, S., Wenzel, J.: Differential cryptanalysis of round-reduced SIMON and SPECK. In: Cid, C., Rechberger, C. (eds.) FSE 2014. LNCS, vol. 8540, pp. 525–545. Springer, Heidelberg (2015). https://doi.org/10.1007/978-3-662-46706-0_27
Alkhzaimi, H.A., Lauridsen, M.M.: Cryptanalysis of the SIMON family of block ciphers. Cryptology ePrint Archive, Report 2013/543 (2013). https://eprint.iacr.org/2013/543
Aumasson, J.-P., Fischer, S., Khazaei, S., Meier, W., Rechberger, C.: New features of Latin dances: analysis of Salsa, ChaCha, and Rumba. In: Nyberg, K. (ed.) FSE 2008. LNCS, vol. 5086, pp. 470–488. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-71039-4_30
Bao, Z., Guo, J., Liu, M., Ma, L., Tu, Y.: Enhancing differential-neural cryptanalysis. Cryptology ePrint Archive, Report 2021/719 (2021). https://eprint.iacr.org/2021/719
Beaulieu, R., Shors, D., Smith, J., Treatman-Clark, S., Weeks, B., Wingers, L.: The SIMON and SPECK Families of Lightweight Block Ciphers. Cryptology ePrint Archive, Report 2013/404 (2013). https://eprint.iacr.org/2013/404
Benamira, A., Gerault, D., Peyrin, T., Tan, Q.Q.: A deeper look at machine learning-based cryptanalysis. In: Canteaut, A., Standaert, F.-X. (eds.) EUROCRYPT 2021. LNCS, vol. 12696, pp. 805–835. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-77870-5_28
Biham, E., Chen, R.: Near-collisions of SHA-0. In: Franklin, M. (ed.) CRYPTO 2004. LNCS, vol. 3152, pp. 290–305. Springer, Heidelberg (2004). https://doi.org/10.1007/978-3-540-28628-8_18
Biryukov, A., Roy, A., Velichkov, V.: Differential analysis of block ciphers SIMON and SPECK. In: Cid, C., Rechberger, C. (eds.) FSE 2014. LNCS, vol. 8540, pp. 546–570. Springer, Heidelberg (2015). https://doi.org/10.1007/978-3-662-46706-0_28
De Cannière, C., Rechberger, C.: Finding SHA-1 characteristics: general results and applications. In: Lai, X., Chen, K. (eds.) ASIACRYPT 2006. LNCS, vol. 4284, pp. 1–20. Springer, Heidelberg (2006). https://doi.org/10.1007/11935230_1
Dinur, I.: Improved differential cryptanalysis of round-reduced speck. In: Joux, A., Youssef, A. (eds.) SAC 2014. LNCS, vol. 8781, pp. 147–164. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-13051-4_9
Gohr, A.: Implementation of the Improving Attacks on Round-Reduced Speck32/64 Using Deep Learning. GitHub Repository (2019). https://github.com/agohr/deep_speck
Gohr, A.: improving attacks on round-reduced speck32/64 using deep learning. In: Boldyreva, A., Micciancio, D. (eds.) CRYPTO 2019. LNCS, vol. 11693, pp. 150–179. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-26951-7_6
Gomez, A.N., Huang, S., Zhang, I., Li, B.M., Osama, M. and Kaiser, L.: Unsupervised cipher cracking using discrete gans. In: 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, 30 April–3 May 2018, Conference Track Proceedings (2018). OpenReview.net
Grassi, L.: Mixture differential cryptanalysis: a new approach to distinguishers and attacks on round-reduced AES. IACR Trans. Symmetric Cryptol. 2018(2), 133–160 (2018)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, NV, USA, 27–30 June 2016, pp. 770–778. IEEE Computer Society (2016)
Hu, J., Shen, L., Albanie, S., Sun, G., Wu, E.: Squeeze-and-excitation networks. IEEE Trans. Pattern Anal. Mach. Intell. 42(8), 2011–2023 (2020)
Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, 21–26 July 2017, pp. 2261–2269. IEEE Computer Society (2017)
Joux, A., Peyrin, T.: Hash functions and the (Amplified) Boomerang attack. In: Menezes, A. (ed.) CRYPTO 2007. LNCS, vol. 4622, pp. 244–263. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-74143-5_14
Klima, V.: Tunnels in hash functions: MD5 collisions within a minute. Cryptology ePrint Archive, Report 2006/105 (2006). https://eprint.iacr.org/2006/105
Knellwolf, S., Meier, W., Naya-Plasencia, M.: Conditional differential cryptanalysis of NLFSR-based cryptosystems. In: Abe, M. (ed.) ASIACRYPT 2010. LNCS, vol. 6477, pp. 130–145. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-17373-8_8
Kölbl, S., Leander, G., Tiessen, T.: Observations on the SIMON block cipher family. In: Gennaro, R., Robshaw, M. (eds.) CRYPTO 2015. LNCS, vol. 9215, pp. 161–185. Springer, Heidelberg (2015). https://doi.org/10.1007/978-3-662-47989-6_8
Leurent, G.: Construction of differential characteristics in ARX designs application to skein. In: Canetti, R., Garay, J.A. (eds.) CRYPTO 2013. LNCS, vol. 8042, pp. 241–258. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-40041-4_14
Rijsdijk, J., Wu, L., Perin, G., Picek, S.: Reinforcement learning for hyperparameter tuning in deep learning-based side-channel analysis. IACR TCHES 2021(3), 677–707 (2021). https://tches.iacr.org/index.php/TCHES/article/view/8989
Rivest, R.L.: Cryptography and machine learning. In: Imai, H., Rivest, R.L., Matsumoto, T. (eds.) ASIACRYPT 1991. LNCS, vol. 739, pp. 427–439. Springer, Heidelberg (1993). https://doi.org/10.1007/3-540-57332-1_36
Silver, D., et al.: A general reinforcement learning algorithm that masters chess, shogi, and go through self-play. Science 362(6419), 1140–1144 (2018)
Song, L., Huang, Z., Yang, Q.: Automatic differential analysis of ARX block ciphers with application to SPECK and LEA. Cryptology ePrint Archive, Report 2016/209 (2016). https://eprint.iacr.org/2016/209
Wang, N., Wang, X., Jia, K., Zhao, J.: Differential attacks on reduced SIMON versions with dynamic key-guessing techniques. Cryptology ePrint Archive, Report 2014/448 (2014). https://eprint.iacr.org/2014/448
Wang, X., Yu, H.: How to break MD5 and other hash functions. In: Cramer, R. (ed.) EUROCRYPT 2005. LNCS, vol. 3494, pp. 19–35. Springer, Heidelberg (2005). https://doi.org/10.1007/11426639_2
Acknowledgments
The authors would like to thank anonymous reviewers for their insightful and helpful comments which helped us improve the manuscript significantly. This research is partially supported by Nanyang Technological University in Singapore under Start-up Grant 04INS000397C230, and Ministry of Education in Singapore under Grants RG91/20 and MOE2019-T2-1-060; Zhenzhen Bao was supported by the Gopalakrishnan – NTU Presidential Postdoctoral Fellowship 2020; the Tsinghua University in China under Start-up Grant 533344001; the National Key R &D Program of China (Grant No. 2018YFA0704701), the Major Program of Guangdong Basic and Applied Research (Grant No. 2019B030302008), the Shandong Province Key R &D Project (Nos. 2020ZLYS09 and 2019JZZY010133). Meicheng Liu was supported by the National Natural Science Foundation of China (Grant Nos. 62122085 and 12231015), and the Youth Innovation Promotion Association of Chinese Academy of Sciences.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 International Association for Cryptologic Research
About this paper
Cite this paper
Bao, Z., Guo, J., Liu, M., Ma, L., Tu, Y. (2022). Enhancing Differential-Neural Cryptanalysis. In: Agrawal, S., Lin, D. (eds) Advances in Cryptology – ASIACRYPT 2022. ASIACRYPT 2022. Lecture Notes in Computer Science, vol 13791. Springer, Cham. https://doi.org/10.1007/978-3-031-22963-3_11
Download citation
DOI: https://doi.org/10.1007/978-3-031-22963-3_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-22962-6
Online ISBN: 978-3-031-22963-3
eBook Packages: Computer ScienceComputer Science (R0)