Explainable machine learning frameworks for cryptography protocol design using discrete structures

Namdev, Mayank; Showkat, Katib; Vishnoi, Susheela; Dadheech, Pankaj

doi:10.47974/JDMSC-2423

Open Access Research Article

Explainable machine learning frameworks for cryptography protocol design using discrete structures

Mayank Namdevmayank.namdev@jaipur.manipal.eduDepartment of Computer Science and Engineering Manipal University JaipurJaipur, Rajasthan, 303007, IndiaView full profile → , Katib Showkatkatibzargar@gmail.comDepartment of Computer Science and Engineering Vivekananda Global UniversityJaipur, Rajasthan, 303012, IndiaView full profile → , ^*Susheela VishnoiCorresponding authorsusheela.vishnoi@jaipur.manipal.eduDepartment of Computer Science and Engineering Manipal University JaipurJaipur, Rajashtan, 302012, IndiaView full profile → , Pankaj Dadheechpankajdadheech777@gmail.comDepartment of Computer Science and Engineering Swami Keshvanand Institute of Technology, Management & Gramothan JaipurDepartment of Computer Science & Engineering Swami Keshvanand Institute of Technology, Management & GramothanJaipur, Rajasthan, 302017, IndiaView full profile →

* Corresponding author · click or hover a name for details

pp. 2059–2069Vol. 28Issue 5-BAugust 2025DOI: 10.47974/JDMSC-2423 Crossmark XML

Received:: 06 Nov 2024
Published Online:: 30 Aug 2025
Article type:: Research Article
Language:: EN
Article no.:: JDMSC-2423
Pages:: 2059–2069

Abstract

This paper introduces Explainable Secure-Net, a new framework that leverages machine learning to aid in the automatic design of cryptographic key-exchange protocols whilst providing transparent human-readable explanations for each decision made. Classical techniques are based on expert-designed algebraic rules and usually lead to long and intricate manual proofs, whereas Explainable Secure-Net encodes some core discrete-mathematical ingredients e.g., finite-field operations and protocol graph structures into a graph neural network. For each step of the protocol that the model proposes, a light-weight explanation module serves to highlight what drove its choice with respect to distinguishing features, keeping the design process transparent and readily auditable. For security, all candidate protocols are run through a hybrid verification process which marries formal symbolic verification to large scale statistical testing. On a 1,500 benchmarks of synthetic protocol examples, Explainable Secure-Net achieves 94 % synthesis accuracy and explores more than 95 % of the potential key-exchange space, while keeping the false-negative vulnerability rate below 1 % and providing 128-bit security guarantees. Our findings show that it is feasible to speed protocol innovation without a loss in terms of rigor and interpretability. We argue that Explainable Secure-Net is an important first step towards machine-augmented cryptographic design tools that can automatically propose, explain, and certify secure protocol definitions for a variety of cryptographic tasks.

Keywords

AI-augmented cryptanalysis Reinforcement learning Combinatorial key-search Block-cipher security Search-space pruning Automated key recovery

Subject Classifications

68T01

References

[1] I. Ryffel and J. K. B. Wieder, “Cryptography for Privacy-Preserving Machine Learning,” IEEE Security & Privacy, vol. 19, no. 3, pp. 81–89 (May–Jun. 2021).
[2] W. Shah and M. Stibor, “Hybrid Neural-Cryptography: Instant Encryption via Deep Autoencoders,” Neural Computing and Applications, vol. 36, no. 8, pp. 7411–7423 (Aug. 2024).
[3] A. Krishnaa, “Inner Magic and Inner Antimagic Graphs in Cryptography,” Journal of Discrete Mathematical Sciences and Cryptography, vol. 22, no. 6, pp. 1057–1066 (2019).
[4] H. Zhang, B. Wu, X. Yuan, S. Pan, H. Tong, and J. Pei, “Trustworthy Graph Neural Networks: Aspects, Methods, and Trends,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 45, no. 6, pp. 3200–3220 (Jun. 2023).
[5] J. Kakkad, J. Jannu, K. Sharma, C. Aggarwal, and S. Medya, “A Survey on Explainability of Graph Neural Networks,” ACM Comput. Surveys, vol. 55, no. 4, Art. 83 (Sep. 2023).
[6] I. Ryffel and J. K. B. Wieder, “Cryptography for Privacy-Preserving Machine Learning,” IEEE Security & Privacy, vol. 19, no. 3, pp. 81–89 (May–Jun. 2021).
[7] P. Kumar, A. Yadav, and R. Singh, “Neural-enhanced metaheuristics for secure key scheduling in lightweight cryptography,” Journal of Information & Optimization Sciences, vol. 47, no. 4, pp. 865–882 (Apr. 2025).
[8] N. Sharma, V. Bansal, and T. Rathi, “Graph-theoretic deep reinforcement learning for optimization of communication protocols,” Journal of Interdisciplinary Mathematics, vol. 29, no. 2, pp. 215–232 (Mar. 2025).
[9] H. Zhang, B. Wu, X. Yuan, S. Pan, H. Tong, and J. Pei, “Trustworthy Graph Neural Networks: Aspects, Methods, and Trends,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 45, no. 6, pp. 3200–3220 (Jun. 2023).

Views: 152Downloads: 5Citations: 0