Automated Graph Optimization for Cryptographic Neural Network Inference

Authors

  • Victor Manuel Sabater Department of Computer Science, Colorado State University, Fort Collins, CO, USA. Author
  • Madhavi L. Ramanathan Department of Computer Science, Binghamton University, Binghamton, NY, USA. Author
  • Rhianne MacCool Department of Computer Science, University of New Hampshire, Durham, NH, USA. Author
  • April D. Kerr School of Information Technology, University of Cincinnati, Cincinnati, OH, USA. Author

Keywords:

cryptographic neural network inference, graph optimization, homomorphic encryption, secure multiparty computation, privacy-enhancing technologies, systems architecture, governance, sustainability

Abstract

Automated graph optimization for cryptographic neural network inference has emerged as a critical systems challenge at the intersection of privacy-enhancing cryptography, machine learning compilers, and distributed infrastructure. This paper examines how compiler-level transformations can reconcile the competing demands of cryptographic correctness, latency, throughput, memory pressure, energy consumption, and deployment governance. It argues that graph optimization for encrypted inference cannot be treated as a narrow arithmetic rewrite problem; rather, it is a socio-technical infrastructure problem in which optimization choices redistribute trust, cost, risk, and accountability across cloud providers, model owners, data subjects, regulators, and hardware vendors. The discussion covers homomorphic encryption, secure multiparty computation, hybrid protocols, and hardware acceleration, with attention to architectural trade-offs in optimization pipelines, verification and robustness, sustainability, fairness, and policy. Case illustrations from private inference services and edge deployments show that automated graph optimization can improve feasibility, but only when paired with transparent governance, reproducible benchmarking, and security validation. The paper concludes with a research agenda for composable, auditable, and sustainable cryptographic inference graphs.

References

1. Gentry, C. (2009). Fully homomorphic encryption using ideal lattices. In Proceedings of the 41st Annual ACM Symposium on Theory of Computing (pp. 169–178). Association for Computing Machinery.

2. 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.

3. Chillotti, I., Gama, N., Georgieva, M., & Izabachène, M. (2020). TFHE: Fast fully homomorphic encryption over the torus. Journal of Cryptology, 33(1), 34–91.

4. Dathathri, R., Saarikivi, O., Chen, H., Laine, K., Lauter, K., Maleki, S., & Musuvathi, M. (2019). CHET: An optimizing compiler for fully homomorphic neural-network inference. In Proceedings of the 40th ACM SIGPLAN Conference on Programming Language Design and Implementation (pp. 1–16). Association for Computing Machinery.

5. Juvekar, C., Vaikuntanathan, V., & Chandrakasan, A. (2018). GAZELLE: A low latency framework for secure neural network inference. In Proceedings of the 27th USENIX Security Symposium (pp. 1651–1669). USENIX Association.

6. 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). Association for Computing Machinery.

7. Mohassel, P., & Zhang, Y. (2017). SecureML: A system for scalable privacy-preserving machine learning. In 2017 IEEE Symposium on Security and Privacy (pp. 19–38). IEEE.

8. 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 on Asia Conference on Computer and Communications Security (pp. 707–721). Association for Computing Machinery.

9. Mishra, P., Lehmkuhl, R., Srinivasan, A., Zheng, W., & Popa, R. A. (2020). Delphi: A cryptographic inference service for neural networks. In Proceedings of the 29th USENIX Security Symposium (pp. 2505–2522). USENIX Association.

10. Rathee, D., Rathee, M., Kumar, N., Chandran, N., Gupta, D., Rastogi, A., & Sharma, R. (2020). CrypTFlow2: Practical 2-party secure inference. In Proceedings of the 2020 ACM SIGSAC Conference on Computer and Communications Security (pp. 325–342). Association for Computing Machinery.

11. Knott, B., Venkataraman, S., Hannun, A., Sengupta, S., Ibrahim, M., & van der Maaten, L. (2021). CrypTen: Secure multi-party computation meets machine learning. In Advances in Neural Information Processing Systems 34 (pp. 4961–4973). Curran Associates.

12. Lou, Q., & Jiang, L. (2019). SHE: A fast and accurate deep neural network for encrypted data. In Advances in Neural Information Processing Systems 32 (pp. 10057–10065). Curran Associates.

13. Benaissa, A., Retiat, B., Cebere, B., & Belfiore, J.-C. (2021). TenSEAL: A library for encrypted tensor operations using homomorphic encryption. arXiv. https://arxiv.org/abs/2104.03152

14. Boemer, F., Costache, A., Cammarota, R., & Wierzynski, C. (2019). nGraph-HE: A graph compiler for deep learning on homomorphically encrypted data. In Proceedings of the 16th ACM International Conference on Computing Frontiers (pp. 3–13). Association for Computing Machinery.

15. Sanyal, A., Kusner, M. J., Gascón, A., & Kanade, V. (2018). TAPAS: Tricks to accelerate (encrypted) prediction as a service. In Proceedings of the 35th International Conference on Machine Learning (pp. 4490–4499). PMLR.

16. Peng, H., Huang, S., Zhou, T., Luo, Y., Wang, C., Wang, Z., ... & Ding, C. (2023, October). Autorep: Automatic relu replacement for fast private network inference. In 2023 IEEE/CVF International Conference on Computer Vision (ICCV) (pp. 5155-5165). IEEE.

17. Watson, J.-L., Wagh, S., & Popa, R. A. (2022). Piranha: A GPU platform for secure computation. In Proceedings of the 31st USENIX Security Symposium (pp. 827–844). USENIX Association.

18. Acar, A., Aksu, H., Uluagac, A. S., & Conti, M. (2018). A survey on homomorphic encryption schemes: Theory and implementation. ACM Computing Surveys, 51(4), 1–35.

19. Chen, T., Moreau, T., Jiang, Z., Zheng, L., Yan, E., Cowan, M., ... & Ceze, L. (2018). TVM: An automated end-to-end optimizing compiler for deep learning. In Proceedings of the 13th USENIX Symposium on Operating Systems Design and Implementation (pp. 578–594). USENIX Association.

20. Cramer, R., Damgård, I., & Nielsen, J. B. (2015). Secure multiparty computation and secret sharing. Cambridge University Press.

Downloads

Published

2026-10-10