Cheetah: Optimizing and Accelerating Homomorphic Encryption for Private Inference
By Brandon Reagen, Wooseok Choi, Yeongil Ko, Vincent T. Lee, Hsien-Hsin S. Lee, Gu-Yeon Wei, David M. Brooks. 2020
187 citations recorded by the source corpus.
As the application of deep learning continues to grow, so does the amount of data used to make predictions. While traditionally big-data deep learning was constrained by computing performance and off-chip memory bandwidth, a new constraint has emerged: privacy. One solution is homomorphic encryption (HE). Applying HE to the client-cloud model allows cloud services to perform inferences directly on clients’ encrypted data. While HE can meet privacy constraints it introduces enormous computational challenges and remains impractically slow on current systems.This paper introduces Cheetah, a set of algorithmic and hardware optimizations for server-side HE DNN inference. Cheetah proposes HE-parameter tuning and operator scheduling optimizations, which together deliver up to $79 \times$ speedup over the state-of-the-art. However, HE inference still falls short of real-time inference speeds by nearly four orders of magnitude. Cheetah further proposes an accelerator architecture to understand the degree of speedup hardware can provide and whether it can bridge HE’s real-time performance gap. We evaluate several DNNs and find that privacy-preserving HE inference for ResNet50 can approach real-time speeds with a 587mm2 accelerator dissipating 30W in 5nm.
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Abhishek Singh
- Praneeth Vepakomma
- Ramesh Raskar
- Fatemehsadat Mirshghallah
- Hadi Esmaeilzadeh
- Mohammadkazem Taram
- Sang-Min Park
- Woosuk Choi
Citations in the corpus
- Privacy in Deep Learning: A Survey (2020, background): “1] # # #MNIST, Faces, Cancer, DiabetesImage Classification w/ DNN FHE–DiNN [132] # # #MNIST Image Classification w/ DNN Face Match [133] # # #LFW, IJB-A, IJB-B, CASIA Face recognition with CNNs Cheetah [134] # # #MNIST, Imagenet Image Classification w/ DNN EPIC [119] # # #CIFAR-10, MIT, Caltech Image C”
- Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network (2021, methodology): “Hybrid model with FHE and MPC Some previous works evaluate the non-arithmetic activation functions with the multiparty computation technique to implement the standard well-known machine learning model preserving privacy [20, 21, 22, 23, 15].”
- A Metaverse: Taxonomy, Components, Applications, and Open Challenges (2022, methodology): “[340] introduced Cheetah, a set of algorithms and hardware optimizations for server-side HE DNN inference to approach”