mlsanityData to 7 Oct 2026

Faster CryptoNets: Leveraging Sparsity for Real-World Encrypted Inference

arXiv 1811.09953 ↗First posted 25 Nov 2018cs.CRcited by 45

Not on the map. Why?

Abstract (arXiv)

Homomorphic encryption enables arbitrary computation over data while it remains encrypted. This privacy-preserving feature is attractive for machine learning, but requires significant computational time due to the large overhead of the encryption scheme. We present Faster CryptoNets, a method for efficient encrypted inference using neural networks. We develop a pruning and quantization approach that leverages sparse representations in the underlying cryptosystem to accelerate inference. We derive an optimal approximation for popular activation functions that achieves maximally-sparse encodings and minimizes approximation error. We also show how privacy-safe training techniques can be used to reduce the overhead of encrypted inference for real-world datasets by leveraging transfer learning and differential privacy. Our experiments show that our method maintains competitive accuracy and achieves a significant speedup over previous methods. This work increases the viability of deep learning systems that use homomorphic encryption to protect user privacy.

Key facts. Cited by 45 papers in the map (data through 7 Oct 2026). First posted on arXiv 25 Nov 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 7; 2022: 10; 2023: 10; 2024: 10; 2025: 4; 2026: 4.

45 papers.

Sort:

  1. 2023Offsite-Tuning: Transfer Learning without Full Modelcited by 77
  2. 2022MPCFormer: fast, performant and private Transformer inference with MPCcited by 67
  3. 2021Privacy-Preserving Machine Learning with Fully Homomorphic Encryption for Deep Neural Networkcited by 57
  4. 2022QuadraLib: A Performant Quadratic Neural Network Library for Architecture Optimization and Design Explorationcited by 21
  5. 2021Circa: Stochastic ReLUs for Private Deep Learningcited by 16
  6. 2023AutoFHE: Automated Adaption of CNNs for Efficient Evaluation over FHEcited by 15
  7. 2021Sisyphus: A Cautionary Tale of Using Low-Degree Polynomial Activations in Privacy-Preserving Deep Learningcited by 14
  8. 2022FedVLN: Privacy-preserving Federated Vision-and-Language Navigationcited by 8
  9. 2022Deploying Convolutional Networks on Untrusted Platforms Using 2D Holographic Reduced Representationscited by 7
  10. 2024Ditto: Quantization-aware Secure Inference of Transformers upon MPCcited by 7
  11. 2024Comet: A Communication-efficient and Performant Approximation for Private Transformer Inferencecited by 7
  12. 2021Enabling Homomorphically Encrypted Inference for Large DNN Modelscited by 6
33 more
  1. 2022Private and Reliable Neural Network Inferencecited by 6
  2. 2022Volley Revolver: A Novel Matrix-Encoding Method for Privacy-Preserving Neural Networks (Inference)cited by 5
  3. 2024MPC-Minimized Secure LLM Inferencecited by 5
  4. 2022Audit and Improve Robustness of Private Neural Networks on Encrypted Datacited by 4
  5. 2023TT-TFHE: a Torus Fully Homomorphic Encryption-Friendly Neural Network Architecturecited by 4
  6. 2023Optimized Layerwise Approximation for Efficient Private Inference on Fully Homomorphic Encryptioncited by 4
  7. 2021SoK: Privacy-preserving Deep Learning with Homomorphic Encryptioncited by 2
  8. 2022Efficient Pruning for Machine Learning Under Homomorphic Encryptioncited by 2
  9. 2023Dash: Accelerating Distributed Private Convolutional Neural Network Inference with Arithmetic Garbled Circuitscited by 2
  10. 2023Hyperdimensional Computing as a Rescue for Efficient Privacy-Preserving Machine Learning-as-a-Servicecited by 2
  11. 2024From Algorithm to Hardware: A Survey on Efficient and Safe Deployment of Deep Neural Networkscited by 2
  12. 2025CryptoFace: End-to-End Encrypted Face Recognitioncited by 2
  13. 2021Efficient CNN Building Blocks for Encrypted Datacited by 1
  14. 2021FFConv: Fast Factorized Convolutional Neural Network Inference on Encrypted Datacited by 1
  15. 2022Partially Oblivious Neural Network Inferencecited by 1
  16. 2023Privacy-Preserving 3-Layer Neural Network Trainingcited by 1
  17. 2023SelectFormer: Private and Practical Data Selection for Transformerscited by 1
  18. 2024Bi-CryptoNets: Leveraging Different-Level Privacy for Encrypted Inferencecited by 1
  19. 2024Neural Networks with (Low-Precision) Polynomial Approximations: New Insights and Techniques for Accuracy Improvementcited by 1
  20. 2024DCT-CryptoNets: Scaling Private Inference in the Frequency Domaincited by 1
  21. 2024SoK: Towards Security and Safety of Edge AIcited by 1
  22. 2022MixNN: A design for protecting deep learning modelscited by 0
  23. 2023ArctyrEX : Accelerated Encrypted Execution of General-Purpose Applicationscited by 0
  24. 2023Integrating Homomorphic Encryption and Trusted Execution Technology for Autonomous and Confidential Model Refining in Cloudcited by 0
  25. 2024I can't see it but I can Fine-tune it: On Encrypted Fine-tuning of Transformers using Fully Homomorphic Encryptioncited by 0
  26. 2024Efficient Privacy-Preserving KAN Inference Using Homomorphic Encryptioncited by 0
  27. 2025SecPE: Secure Prompt Ensembling for Private and Robust Large Language Modelscited by 0
  28. 2025EP-HDC: Hyperdimensional Computing with Encrypted Parameters for High-Throughput Privacy-Preserving Inferencecited by 0
  29. 2025PRISM: Privacy-preserving Inference System with Homomorphic Encryption and Modular Activationcited by 0
  30. 2026RobPI: Robust Private Inference against Malicious Clientcited by 0
  31. 2026Towards Deep Encrypted Training: Low-Latency, Memory-Efficient, and High-Throughput Inference for Privacy-Preserving Neural Networkscited by 0
  32. 2026Privatar: Scalable Privacy-preserving Multi-user VR via Secure Offloadingcited by 0
  33. 2026Encrypt What Matters: When Selective Homomorphic Inference Is Efficientcited by 0