CHE: Channel-Wise Homomorphic Encryption for Ciphertext Inference in Convolutional Neural Network
Privacy-preserving deep learning (PPDL), which leverages Homomorphic Encryption (HE), has attracted attention as a promising approach to ensure the privacy of deep learning applications’ data. While recent studies have developed and evaluated the HE-based PPDL algorithms, the achieved per...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9903645/ |
| _version_ | 1811182536098840576 |
|---|---|
| author | Tianying Xie Hayato Yamana Tatsuya Mori |
| author_facet | Tianying Xie Hayato Yamana Tatsuya Mori |
| author_sort | Tianying Xie |
| collection | DOAJ |
| description | Privacy-preserving deep learning (PPDL), which leverages Homomorphic Encryption (HE), has attracted attention as a promising approach to ensure the privacy of deep learning applications’ data. While recent studies have developed and evaluated the HE-based PPDL algorithms, the achieved performances, such as accuracy and latency, need improvement to make the applications practical. This work aims to improve the performance of the image classification of HE-based PPDL by combining two approaches — Channel-wise Homomorphic Encryption (CHE) and Batch Normalization (BN) with coefficient merging. Although these are commonly used schemes, their detailed algorithms and formulations have not been clearly described. The main contribution of the current study is to provide complete and reproducible descriptions of these schemes. We evaluate our CHE and BN implementation by targeting the Cheon-Kim-Kim-Song scheme as an HE scheme and Convolution Neural Network (CNN) as a machine learning scheme while using the MNIST and CIFAR-10 as the datasets. In addition, we compare the results with the five state-of-the-art neural network architectures. Our experiments demonstrate that the CHE can serve as a tool for empirically achieving shorter latency (the shortest 7.76 seconds) and higher accuracy (the highest 99.32%) compared with the previous studies that aimed to establish the classification of the encrypted MNIST data with CNN. Our approach can aid in designing a more robust and flexible PPDL. |
| first_indexed | 2024-04-11T09:33:16Z |
| format | Article |
| id | doaj.art-606be5d27e5c40309e425910451085fe |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-11T09:33:16Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-606be5d27e5c40309e425910451085fe2022-12-22T04:31:47ZengIEEEIEEE Access2169-35362022-01-011010744610745810.1109/ACCESS.2022.32101349903645CHE: Channel-Wise Homomorphic Encryption for Ciphertext Inference in Convolutional Neural NetworkTianying Xie0https://orcid.org/0000-0002-9597-841XHayato Yamana1https://orcid.org/0000-0001-7542-4826Tatsuya Mori2https://orcid.org/0000-0003-1583-4174Department of Computer Science and Communications Engineering, Waseda University, Tokyo, JapanDepartment of Computer Science and Communications Engineering, Waseda University, Tokyo, JapanDepartment of Computer Science and Communications Engineering, Waseda University, Tokyo, JapanPrivacy-preserving deep learning (PPDL), which leverages Homomorphic Encryption (HE), has attracted attention as a promising approach to ensure the privacy of deep learning applications’ data. While recent studies have developed and evaluated the HE-based PPDL algorithms, the achieved performances, such as accuracy and latency, need improvement to make the applications practical. This work aims to improve the performance of the image classification of HE-based PPDL by combining two approaches — Channel-wise Homomorphic Encryption (CHE) and Batch Normalization (BN) with coefficient merging. Although these are commonly used schemes, their detailed algorithms and formulations have not been clearly described. The main contribution of the current study is to provide complete and reproducible descriptions of these schemes. We evaluate our CHE and BN implementation by targeting the Cheon-Kim-Kim-Song scheme as an HE scheme and Convolution Neural Network (CNN) as a machine learning scheme while using the MNIST and CIFAR-10 as the datasets. In addition, we compare the results with the five state-of-the-art neural network architectures. Our experiments demonstrate that the CHE can serve as a tool for empirically achieving shorter latency (the shortest 7.76 seconds) and higher accuracy (the highest 99.32%) compared with the previous studies that aimed to establish the classification of the encrypted MNIST data with CNN. Our approach can aid in designing a more robust and flexible PPDL.https://ieeexplore.ieee.org/document/9903645/Channel-wisehomomorphic encryptionCNNprivacy-preservingciphertext inference |
| spellingShingle | Tianying Xie Hayato Yamana Tatsuya Mori CHE: Channel-Wise Homomorphic Encryption for Ciphertext Inference in Convolutional Neural Network IEEE Access Channel-wise homomorphic encryption CNN privacy-preserving ciphertext inference |
| title | CHE: Channel-Wise Homomorphic Encryption for Ciphertext Inference in Convolutional Neural Network |
| title_full | CHE: Channel-Wise Homomorphic Encryption for Ciphertext Inference in Convolutional Neural Network |
| title_fullStr | CHE: Channel-Wise Homomorphic Encryption for Ciphertext Inference in Convolutional Neural Network |
| title_full_unstemmed | CHE: Channel-Wise Homomorphic Encryption for Ciphertext Inference in Convolutional Neural Network |
| title_short | CHE: Channel-Wise Homomorphic Encryption for Ciphertext Inference in Convolutional Neural Network |
| title_sort | che channel wise homomorphic encryption for ciphertext inference in convolutional neural network |
| topic | Channel-wise homomorphic encryption CNN privacy-preserving ciphertext inference |
| url | https://ieeexplore.ieee.org/document/9903645/ |
| work_keys_str_mv | AT tianyingxie chechannelwisehomomorphicencryptionforciphertextinferenceinconvolutionalneuralnetwork AT hayatoyamana chechannelwisehomomorphicencryptionforciphertextinferenceinconvolutionalneuralnetwork AT tatsuyamori chechannelwisehomomorphicencryptionforciphertextinferenceinconvolutionalneuralnetwork |