nGraph-HE: a graph compiler for deep learning on homomorphically encrypted data
作者： Fabian Boemer , Yixing Lao , Rosario Cammarota , Casimir Wierzynski
关键词: Ciphertext 、 Graph (abstract data type) 、 Computer science 、 Compiler 、 Plaintext 、 Encryption 、 Homomorphic encryption 、 Constant folding 、 Programming language 、 Cryptography
摘要: Homomorphic encryption (HE)---the ability to perform computation on encrypted data---is an attractive remedy increasing concerns about data privacy in deep learning (DL). However, building DL models that operate ciphertext is currently labor-intensive and requires simultaneous expertise DL, cryptography, software engineering. frameworks recent advances graph compilers have greatly accelerated the training deployment of various computing platforms. We introduce nGraph-HE, extension nGraph, Intel's compiler, which enables trained with popular such as TensorFlow while simply treating HE another hardware target. Our graph-compiler approach HE-aware optimizations- implemented at compile-time, constant folding HE-SIMD packing, run-time, special value plaintext bypass. Furthermore, nGraph-HE integrates TensorFlow, enabling scientists benchmark minimal overhead.
arxiv.org
acm.org
sci-hub.se 参考文章(37)
Mihir Bellare, Viet Tung Hoang, Sriram Keelveedhi, Phillip Rogaway, Efficient Garbling from a Fixed-Key Blockcipher ieee symposium on security and privacy. pp. 478- 492 ,(2013) , 10.1109/SP.2013.39
Yangqing Jia, Evan Shelhamer, Jeff Donahue, Sergey Karayev, Jonathan Long, Ross Girshick, Sergio Guadarrama, Trevor Darrell, Caffe: Convolutional Architecture for Fast Feature Embedding acm multimedia. pp. 675- 678 ,(2014) , 10.1145/2647868.2654889
Tianjun Xiao, Tianqi Chen, Chiyuan Zhang, Zheng Zhang, Yutian Li, Min Lin, Minjie Wang, Naiyan Wang, Mu Li, Bing Xu, MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems arXiv: Distributed, Parallel, and Cluster Computing. ,(2015)
Ran Gilad-Bachrach, Kristin Lauter, John Wernsing, Michael Naehrig, Kim Laine, Nathan Dowlin, CryptoNets: applying neural networks to encrypted data with high throughput and accuracy international conference on machine learning. pp. 201- 210 ,(2016)
Hao Chen, Kim Laine, Rachel Player, Simple Encrypted Arithmetic Library - SEAL v2.1 financial cryptography. pp. 3- 18 ,(2017) , 10.1007/978-3-319-70278-0_1
Jean-Claude Bajard, Julien Eynard, M. Anwar Hasan, Vincent Zucca, A Full RNS Variant of FV Like Somewhat Homomorphic Encryption Schemes selected areas in cryptography. ,vol. 2016, pp. 423- 442 ,(2016) , 10.1007/978-3-319-69453-5_23
Ilaria Chillotti, Nicolas Gama, Mariya Georgieva, Malika Izabachène, Faster Fully Homomorphic Encryption: Bootstrapping in Less Than 0.1 Seconds international conference on the theory and application of cryptology and information security. ,vol. 2016, pp. 3- 33 ,(2016) , 10.1007/978-3-662-53887-6_1
Abbas Acar, Hidayet Aksu, A. Selcuk Uluagac, Mauro Conti, A Survey on Homomorphic Encryption Schemes: Theory and Implementation ACM Computing Surveys. ,vol. 51, pp. 79- ,(2018) , 10.1145/3214303
T Tanja Lange, DJ Daniel Bernstein, Post-quantum cryptography - dealing with the fallout of physics success. IACR Cryptology ePrint Archive. ,vol. 2017, pp. 314- ,(2017)
Paul Voigt, Axel von dem Bussche, The EU General Data Protection Regulation (GDPR) Springer International Publishing. ,(2017) , 10.1007/978-3-319-57959-7