High-Precision Bootstrapping of RNS-CKKS Homomorphic Encryption Using Optimal Minimax Polynomial Approximation and Inverse Sine Function

Lee, Dong Kun; Lee, Eunsang; Lee, Yong-Woo; Kim, Young Sik; No, Jong-Seon

doi:10.1007/978-3-030-77870-5_22

Cited by 64 publications

(44 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the first time, we implement the ResNet-20 model for the CIFAR-10 dataset [12] using the residue number system CKKS (RNS-CKKS) [4] FHE scheme, which is a variant of the CKKS scheme using the SEAL library 3.6.1 version [13], one of the most reliable libraries implementing the RNS-CKKS scheme. In addition, we implement bootstrapping of RNS-CKKS scheme in the SEAL library according to [6]- [10] in order to support a large number of homomorphic operations for a deep neural network, as the SEAL library does not support the bootstrapping operation. ResNets are one of the historic convolutional neural network (CNN) models which enable a very deep neural network with high accuracy for complex datasets such as the CIFAR-10 and the ImageNet.…”

Section: A Our Contributionmentioning

confidence: 99%

“…There are several techniques and we elaborate only the techniques we used to implement. Lee et al [10] proposed that the modular reduction is represented by the composition of several functions…”

Section: Bootstrapping Of Ckks Schemementioning

confidence: 99%

“…The bootstrapping technique transforms a ciphertext that cannot support the homomorphic multiplication further to a fresh ciphertext by extending the levels of the ciphertext [5], [6]. However, the bootstrapping technique has been actively improved in regard to algorithmic time complexity [7]- [9], precision [10], and implementation [11], which make bootstrapping more practical. The PPML model with many layers has to be implemented with the precise and efficient bootstrapping technique in FHE.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network

et al. 2022

Self Cite

View full text Add to dashboard Cite

Fully homomorphic encryption (FHE) is one of the prospective tools for privacy-preserving machine learning (PPML), and several PPML models have been proposed based on various FHE schemes and approaches. Although the FHE schemes are known as suitable tools to implement PPML models, previous PPML models on FHE such as CryptoNet, SEALion, and CryptoDL are limited to only simple and non-standard types of machine learning models. These non-standard machine learning models are not proven efficient and accurate with more practical and advanced datasets. Previous PPML schemes replace non-arithmetic activation functions with simple arithmetic functions instead of adopting approximation methods and do not use bootstrapping, which enables continuous homomorphic evaluations. Thus, they could not use standard activation functions and could not employ a large number of layers. In this work, we firstly implement the standard ResNet-20 model with the RNS-CKKS FHE with bootstrapping and verify the implemented model with the CIFAR-10 dataset and the plaintext model parameters. Instead of replacing the non-arithmetic functions with the simple arithmetic function, we use state-of-the-art approximation methods to evaluate these non-arithmetic functions, such as the ReLU and softmax, with sufficient precision. Further, for the first time, we use the bootstrapping technique of the RNS-CKKS scheme in the proposed model, which enables us to evaluate an arbitrary deep learning model on the encrypted data. We numerically verify that the proposed model with the CIFAR-10 dataset shows 98.43% identical results to the original ResNet-20 model with non-encrypted data. The classification accuracy of the proposed model is 92.43%±2.65%, which is quite close to that of the original ResNet-20 CNN model, 91.89%. It takes about 3 hours for inference on a dual Intel Xeon Platinum 8280 CPU (112 cores) with 172 GB memory. We think that it opens the possibility of applying the FHE to the advanced deep PPML model.

show abstract

Section: A Our Contributionmentioning

confidence: 99%

Section: Bootstrapping Of Ckks Schemementioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, L should be larger than L boot , and having a larger L is beneficial since it requires less frequent bootstrapping ops to execute a HE application with a fixed multiplicative depth. L boot depends on the bootstrapping algorithm and typically ranges from 10 to 20 -larger L boot permits using more precise and faster bootstrapping algorithms but at the cost of more frequent bootstrapping [12], [16], [37], [52]. The bootstrapping algorithm we use in this paper is based on [37] with updates to meet the latest security and precision requirements [12], [19], [54], and has L boot of 19.…”

Section: Multiplicative Level and He Bootstrappingmentioning

confidence: 99%

“…The level of security for the HE scheme is determined by the λ parameter as it determines the minimum logarithmic-time complexity for an attack [19] to deduce the message from a ct without the secret key. In this work, we target λ of 128 bits, similar to recent HE studies [12], [52], [54] and libraries [32], [60]. A prior study, F1 [66], provided a substandard [4] level of security under 80 bits for CKKS bootstrapping and used smaller cts which simplifies the microarchitecture.…”

Section: E Modern Algorithmic Optimizations In Ckks and T Multa/slotmentioning

confidence: 99%

BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption

Kim,

Kim

et al. 2021

Preprint

View full text Add to dashboard Cite

Homomorphic encryption (HE) enables secure offloading of computation to the cloud by providing computation on encrypted data (ciphertexts). HE is based on noisy encryption schemes such that noise accumulates as we apply more computation to the data. The limited number of operations applicable to the data prevents practical applications from exploiting HE. Bootstrapping enables an unlimited number of operations or fully HE (FHE) by refreshing the ciphertext. Unfortunately, bootstrapping requires a significant amount of additional computation and memory bandwidth. Prior works have proposed hardware accelerators for computation primitives of FHE. However, to the best of our knowledge, this is the first to propose a hardware FHE accelerator tailored to support bootstrapping efficiently.In particular, we propose BTS -Bootstrappable, Technologydriven, Secure accelerator architecture for FHE. We identify the challenges of supporting bootstrapping in the accelerator and analyze the off-chip memory bandwidth and computation required. In particular, given the limitations of modern memory technology, we identify the HE parameter sets that are efficient for FHE acceleration. Based on the insights from our analysis, we propose BTS that effectively exploits parallelism innate in HE operations by arranging a massive number of processing elements in a grid. We present the design and microarchitecture of BTS, including the network-on-chip that exploits the deterministic communication pattern. BTS shows 5,556× and 1,306× improved execution time on ResNet-20 and logistic regression over CPU, using 373.6mm 2 chip area and up to 133.8W of power.

show abstract

Transciphering Framework for Approximate Homomorphic Encryption

Cho

Kim

et al. 2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

High-Precision Bootstrapping of RNS-CKKS Homomorphic Encryption Using Optimal Minimax Polynomial Approximation and Inverse Sine Function

Cited by 64 publications

References 19 publications

Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network

Privacy-Preserving Machine Learning With Fully Homomorphic Encryption for Deep Neural Network

BTS: An Accelerator for Bootstrappable Fully Homomorphic Encryption

Transciphering Framework for Approximate Homomorphic Encryption

Contact Info

Product

Resources

About