nGraph-HE: A Graph Compiler for Deep Learning on Homomorphically Encrypted Data

Fabian Boemer, Yixing Lao, Rosario Cammarota, Casimir Wierzynski
10/23/2018
cs.CR

Official code: NervanaSystems/he-transformer (223 stars, Apache-2.0, TensorFlow, last commit 2023-01-03).

Abstract

Homomorphic encryption (HE)---the ability to perform computation on encrypted data---is an attractive remedy to increasing concerns about data privacy in deep learning (DL). However, building DL models that operate on ciphertext is currently labor-intensive and requires simultaneous expertise in DL, cryptography, and software engineering. DL frameworks and recent advances in graph compilers have greatly accelerated the training and deployment of DL models to various computing platforms. We introduce nGraph-HE, an extension of nGraph, Intel's DL graph compiler, which enables deployment of trained models with popular frameworks such as TensorFlow while simply treating HE as another hardware target. Our graph-compiler approach enables HE-aware optimizations-- implemented at compile-time, such as constant folding and HE-SIMD packing, and at run-time, such as special value plaintext bypass. Furthermore, nGraph-HE integrates with DL frameworks such as TensorFlow, enabling data scientists to benchmark DL models with minimal overhead.

View on arXivView PDF

Code Implementations(1)

NervanaSystems/he-transformerOfficial100%
TensorFlowApache-2.0

nGraph-HE: Deep learning with Homomorphic Encryption (HE) through Intel nGraph

22339C++, CMakeApr 9, 20183 years agoApache-2.0
compilerdeep-learninghomomorphic-encryptionngraphprivacy-preserving+2 more

Similar Papers

EVA: An Encrypted Vector Arithmetic Language and Compiler for Efficient Homomorphic Computation

84% match1 repo4.1k2019

Effect of Homomorphic Encryption on the Performance of Training Federated Learning Generative Adversarial Networks

83% match2022

QuanCrypt-FL: Quantized Homomorphic Encryption with Pruning for Secure Federated Learning

82% match2024

Hyperdimensional Computing as a Rescue for Efficient Privacy-Preserving Machine Learning-as-a-Service

82% match2023

AI-Driven Secure Data Sharing: A Trustworthy and Privacy-Preserving Approach

82% match2025

OptiPrime: Optimizing Private Inference through Protocol-Hardware Co-design

81% match2026

Cite this paper

@article{boemer2018ngraphhe,
  title  = {nGraph-HE: A Graph Compiler for Deep Learning on Homomorphically Encrypted Data},
  author = {Fabian Boemer and Yixing Lao and Rosario Cammarota and Casimir Wierzynski},
  year   = {2018},
  eprint = {1810.10121},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1810.10121}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1810.10121.svg)](https://codeofpaper.com/paper/1810.10121)

Discussion