mlsanityData to 7 Oct 2026

Enabling Homomorphically Encrypted Inference for Large DNN Models

arXiv 2103.16139 ↗First posted 30 Mar 2021cs.CR cs.LG cs.PFcited by 6

Not on the map. Why?

Abstract (arXiv)

The proliferation of machine learning services in the last few years has raised data privacy concerns. Homomorphic encryption (HE) enables inference using encrypted data but it incurs 100x-10,000x memory and runtime overheads. Secure deep neural network (DNN) inference using HE is currently limited by computing and memory resources, with frameworks requiring hundreds of gigabytes of DRAM to evaluate small models. To overcome these limitations, in this paper we explore the feasibility of leveraging hybrid memory systems comprised of DRAM and persistent memory. In particular, we explore the recently-released Intel Optane PMem technology and the Intel HE-Transformer nGraph to run large neural networks such as MobileNetV2 (in its largest variant) and ResNet-50 for the first time in the literature. We present an in-depth analysis of the efficiency of the executions with different hardware and software configurations. Our results conclude that DNN inference using HE incurs on friendly access patterns for this memory configuration, yielding efficient executions.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 15 of its 45 references that are in the map. First posted on arXiv 30 Mar 2021.

Builds on

15 of 45 references.

References · 15

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  3. 2018Gazelle: A Low Latency Framework for Secure Neural Network Inferencecited by 193
  4. 2018A generic framework for privacy preserving deep learningcited by 103
  5. 2019CrypTFlow: Secure TensorFlow Inferencecited by 61
  6. 2018Low Latency Privacy Preserving Inferencecited by 57
  7. 2017CryptoDL: Deep Neural Networks over Encrypted Datacited by 56
  8. 2019XONN: XNOR-based Oblivious Deep Neural Network Inferencecited by 51
  9. 2018Faster CryptoNets: Leveraging Sparsity for Real-World Encrypted Inferencecited by 45
  10. 2018nGraph-HE: A Graph Compiler for Deep Learning on Homomorphically Encrypted Datacited by 24
  11. 2018Intel nGraph: An Intermediate Representation, Compiler, and Executor for Deep Learningcited by 22
  12. 2019nGraph-HE2: A High-Throughput Framework for Neural Network Inference on Encrypted Datacited by 21
3 more
  1. 2020ARIANN: Low-Interaction Privacy-Preserving Deep Learning via Function Secret Sharingcited by 20
  2. 2018Private Machine Learning in TensorFlow using Secure Computationcited by 15
  3. 2019SEALion: a Framework for Neural Network Inference on Encrypted Datacited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 1; 2023: 1; 2024: 1; 2025: 1; 2026: 1.

6 papers.

  1. 2021A methodology for training homomorphicencryption friendly neural networkscited by 10
  2. 2023Training Large Scale Polynomial CNNs for E2E Inference over Homomorphic Encryptioncited by 5
  3. 2022Syfer: Neural Obfuscation for Private Data Releasecited by 3
  4. 2025A Scalable Multi-GPU Framework for Encrypted Large-Model Inferencecited by 2
  5. 2024A Multivocal Literature Review on Privacy and Fairness in Federated Learningcited by 0
  6. 2026Privatar: Scalable Privacy-preserving Multi-user VR via Secure Offloadingcited by 0