EDLaaS:Fully Homomorphic Encryption over Neural Network Graphs for Vision and Private Strawberry Yield Forecasting

被引:2
作者
Onoufriou, George [1 ]
Hanheide, Marc [1 ]
Leontidis, Georgios [2 ]
机构
[1] Univ Lincoln, Sch Comp Sci, Lincoln LN6 7TS, England
[2] Univ Aberdeen, Interdisciplinary Ctr Data &AI & Sch Nat & Comp S, Aberdeen AB24 3FX, Scotland
基金
英国生物技术与生命科学研究理事会;
关键词
fully homomorphic encryption; deep learning; machine learning; privacy-preserving technologies; agri-food; data sharing;
D O I
10.3390/s22218124
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
We present automatically parameterised Fully Homomorphic Encryption (FHE) for encrypted neural network inference and exemplify our inference over FHE-compatible neural networks with our own open-source framework and reproducible examples. We use the fourth generation Cheon, Kim, Kim, and Song (CKKS) FHE scheme over fixed points provided by the Microsoft Simple Encrypted Arithmetic Library (MS-SEAL). We significantly enhance the usability and applicability of FHE in deep learning contexts, with a focus on the constituent graphs, traversal, and optimisation. We find that FHE is not a panacea for all privacy-preserving machine learning (PPML) problems and that certain limitations still remain, such as model training. However, we also find that in certain contexts FHE is well-suited for computing completely private predictions with neural networks. The ability to privately compute sensitive problems more easily while lowering the barriers to entry can allow otherwise too-sensitive fields to begin advantaging themselves of performant third-party neural networks. Lastly, we show how encrypted deep learning can be applied to a sensitive real-world problem in agri-food, i.e., strawberry yield forecasting, demonstrating competitive performance. We argue that the adoption of encrypted deep learning methods at scale could allow for a greater adoption of deep learning methodologies where privacy concerns exist, hence having a large positive potential impact within the agri-food sector and its journey to net zero.
引用
收藏
页数:24
相关论文
共 38 条
  • [1] OpenFHE: Open-Source Fully Homomorphic Encryption Library
    Al Badawi, Ahmad
    Bates, Jack
    Bergamaschi, Flavio
    Cousins, David Bruce
    Erabelli, Saroja
    Genise, Nicholas
    Halevi, Shai
    Hunt, Hamish
    Kim, Andrey
    Lee, Yongwoo
    Liu, Zeyu
    Micciancio, Daniele
    Quah, Ian
    Polyakov, Yuriy
    Saraswathy, R. V.
    Rohloff, Kurt
    Saylor, Jonathan
    Suponitsky, Dmitriy
    Triplett, Matthew
    Vaikuntanathan, Vinod
    Zucca, Vincent
    [J]. PROCEEDINGS OF THE 10TH WORKSHOP ON ENCRYPTED COMPUTING & APPLIED HOMOMORPHIC CRYPTOGRAPHY, WAHC 2022, 2022, : 53 - 63
  • [2] [Anonymous], 2020, Microsoft SEAL (release 3.6)
  • [3] Exponentiating in Pairing Groups
    Bos, Joppe W.
    Costello, Craig
    Naehrig, Michael
    [J]. SELECTED AREAS IN CRYPTOGRAPHY - SAC 2013, 2014, 8282 : 438 - 455
  • [4] Brakerski Z., 2011, 2011277 CRYPT EPRINT
  • [5] Efficient Fully Homomorphic Encryption from (Standard) LWE
    Brakerski, Zvika
    Vaikuntanathan, Vinod
    [J]. 2011 IEEE 52ND ANNUAL SYMPOSIUM ON FOUNDATIONS OF COMPUTER SCIENCE (FOCS 2011), 2011, : 97 - 106
  • [6] Chen Z., 2021, SEAL PYTHON BINDINGS
  • [7] Homomorphic Encryption for Arithmetic of Approximate Numbers
    Cheon, Jung Hee
    Kim, Andrey
    Kim, Miran
    Song, Yongsoo
    [J]. ADVANCES IN CRYPTOLOGY - ASIACRYPT 2017, PT I, 2017, 10624 : 409 - 437
  • [8] Bootstrapping for Approximate Homomorphic Encryption
    Cheon, Jung Hee
    Han, Kyoohyung
    Kim, Andrey
    Kim, Miran
    Song, Yongsoo
    [J]. ADVANCES IN CRYPTOLOGY - EUROCRYPT 2018, PT I, 2018, 10820 : 360 - 384
  • [9] EVA: An Encrypted Vector Arithmetic Language and Compiler for Efficient Homomorphic Computation
    Dathathri, Roshan
    Kostova, Blagovesta
    Saarikivi, Olli
    Dai, Wei
    Laine, Kim
    Musuvathi, Madan
    [J]. PROCEEDINGS OF THE 41ST ACM SIGPLAN CONFERENCE ON PROGRAMMING LANGUAGE DESIGN AND IMPLEMENTATION (PLDI '20), 2020, : 546 - 561
  • [10] DEFRA, 2021, United Kingdom food security report 2021: theme 2: UK food supply sources-GOV