Communication-Efficient Two-Party Secure Computation for Deep Learning Inference
Keywords:
communication-efficient secure computation; two-party computation; private deep learning inference; secure multiparty computation; system architecture; privacy governance; deployment sustainabilityAbstract
Deep learning inference increasingly crosses organizational, jurisdictional, and device boundaries, creating demand for cryptographic methods that protect both client inputs and model parameters. Two-party secure computation offers a principled way to evaluate neural networks without revealing either party's private data, yet practical deployment remains constrained by communication cost, latency instability, infrastructure heterogeneity, and governance complexity. This paper examines communication-efficient two-party secure computation for deep learning inference from a system-level perspective rather than a narrow cryptographic one. It analyzes how protocol choice, network topology, preprocessing strategy, model architecture, and deployment environment interact to shape feasibility, security, and fairness. The discussion considers secure multiparty computation foundations, hybrid protocol composition, homomorphic encryption, differential privacy, and secure aggregation, with attention to their operational trade-offs. It then connects these technical choices to governance, auditability, policy compliance, sustainability, and economic viability. Particular emphasis is placed on the structural tension between reducing communication rounds and preserving strong privacy guarantees under realistic adversarial conditions. The paper also considers robustness, fairness, and redress in private inference systems, arguing that cryptographic protection alone cannot resolve broader questions of accountability and distributive impact. Through cross-domain comparisons in healthcare, finance, mobile computing, and public-sector services, the analysis identifies architectural patterns that may improve deployability without overstating security. The conclusion proposes an interdisciplinary research agenda that integrates protocol design, infrastructure planning, regulatory alignment, and human-centered evaluation for communication-efficient private inference.
References
1. Mohassel, P., & Zhang, Y. (2017). SecureML: A system for scalable privacy-preserving machine learning. In 2017 IEEE Symposium on Security and Privacy (SP) (pp. 19–38). IEEE. https://doi.org/10.1109/SP.2017.12
2. Liu, J., Juuti, M., Lu, Y., & Asokan, N. (2017). Oblivious neural network predictions via MiniONN transformations. In Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security (pp. 619–631). ACM. https://doi.org/10.1145/3133956.3134056
3. Riazi, M. S., Weinert, C., Tkachenko, O., Songhori, E. M., Schneider, T., & Koushanfar, F. (2018). Chameleon: A hybrid secure computation framework for machine learning applications. In Proceedings of the 2018 Asia Conference on Computer and Communications Security (pp. 707–721). ACM. https://doi.org/10.1145/3196494.3196522
4. Juvekar, C., Vaikuntanathan, V., & Chandrakasan, A. (2018). Gazelle: A low latency framework for secure neural network inference. In 27th USENIX Security Symposium (pp. 1651–1669). USENIX Association.
5. Peng, H., Zhou, S., Luo, Y., Xu, N., Duan, S., Ran, R., ... & Ding, C. (2023). Rrnet: Towards relu-reduced neural network for two-party computation based private inference. arXiv preprint arXiv:2302.02292.
6. Yao, A. C. (1986). How to generate and exchange secrets. In 27th Annual Symposium on Foundations of Computer Science (pp. 162–167). IEEE. https://doi.org/10.1109/SFCS.1986.25
7. Goldreich, O., Micali, S., & Wigderson, A. (1987). How to play any mental game. In Proceedings of the nineteenth annual ACM symposium on Theory of computing (pp. 218–229). ACM. https://doi.org/10.1145/28395.28420
8. Beaver, D. (1991). Efficient multiparty protocols using circuit randomization. In Advances in Cryptology—CRYPTO '91 (pp. 420–432). Springer. https://doi.org/10.1007/3-540-46766-1_34
9. Demmler, D., Schneider, T., & Zohner, M. (2015). ABY: A framework for efficient mixed-protocol secure two-party computation. In Proceedings of the 2015 Network and Distributed System Security Symposium. Internet Society. https://doi.org/10.14722/ndss.2015.23113
10. Evans, D., Kolesnikov, V., & Rosulek, M. (2018). A pragmatic introduction to secure multi-party computation. Foundations and Trends in Privacy and Security, 2(2-3), 70–246. https://doi.org/10.1561/3300000017
11. Canetti, R. (2000). Security and composition of multiparty cryptographic protocols. Journal of Cryptology, 13(1), 143–202. https://doi.org/10.1007/s001459910006
12. Dwork, C. (2006). Differential privacy. In Automata, Languages and Programming (pp. 1–12). Springer. https://doi.org/10.1007/11787006_1
13. Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016). Deep learning with differential privacy. In Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security (pp. 308–318). ACM. https://doi.org/10.1145/2976749.2978318
14. Gilad-Bachrach, R., Dowlin, N., Laine, K., Lauter, K., Naehrig, M., & Wernsing, J. (2016). CryptoNets: Applying neural networks to encrypted data with high throughput and accuracy. In Proceedings of the 33rd International Conference on Machine Learning (pp. 201–210). PMLR.
15. Zhang, C., Li, S., Xia, J., Wang, W., Yan, F., & Liu, Y. (2020). BatchCrypt: Efficient homomorphic encryption for cross-silo federated learning. In 2020 USENIX Annual Technical Conference (pp. 493–506). USENIX Association.
16. Kairouz, P., McMahan, H. B., Avent, B., Bellet, A., Bennis, M., Bhagoji, A. N., ... & Zhao, S. (2021). Advances and open problems in federated learning. Foundations and Trends in Machine Learning, 14(1–2), 1–210. https://doi.org/10.1561/2200000083
17. Barocas, S., Hardt, M., & Narayanan, A. (2019). Fairness and machine learning: Limitations and opportunities. fairmlbook.org.
18. Wieringa, R. J. (2014). Design science methodology for information systems and software engineering. Springer. https://doi.org/10.1007/978-3-662-43839-8
19. Zhu, L., Liu, Z., & Han, S. (2019). Deep leakage from gradients. In Advances in Neural Information Processing Systems 32 (pp. 14774–14784). Curran Associates.
20. Mishra, P., Lehmkuhl, R., Srinivasan, A., Zheng, W., & Popa, R. A. (2020). Delphi: A cryptographic inference service for neural networks. In 29th USENIX Security Symposium (pp. 2505–2522). USENIX Association.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Computer Science and Engineering Transactions
This work is licensed under a Creative Commons Attribution 4.0 International License.