Ying Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ying Huang has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Guoqiang Deng
Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation, Center for Applied Mathematics of Guangxi, Guilin University of Electronic Technology (GUET)
3 shared publications - Min Tang
School of Mathematics and Computing Science, Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation, Guilin University of Electronic Technology, Guilin, China
2 shared publications - Min Tang
School of Mathematics and Computing Science, Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation, Guilin University of Electronic Technology, Guilin 541004, China
1 shared publication Xuefeng Duan
1 shared publication- Xuefeng Duan
College of Mathematics and Computational Science, Guilin University of Electronic Technology, Guilin 541004, China
1 shared publication - Yuhao Zhang
School of Mathematics and Computing Science, Guangxi Colleges and Universities Key Laboratory of Data Analysis and Computation, Guilin University of Electronic Technology, Guilin 541004, China
1 shared publication
Latest Publications
- 8 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet Computing
- Vol. 28 (5) , 47-56
Privacy-Preserving Learning of Prediagnosis Models from Distributed Medical Records
Online Medical Pre-Diagnosis (OMPD) systems based on machine learning will benefit individuals for automatic diagnosis while saving time and costs. The inherent security constraints and distributed setting of patient data form obstacles in constructing reliable OMPD models. Existing privacy-preserving machine learning approaches incur substantial communication and computation cost, and are susceptible to inference attacks. Moreover, to accommodate cryptography primitives, machine learning schemes in the encrypted domain are substituted by approximation models, leading to accuracy loss. To mitigate these problems, we build PriD , an efficient and accurate model for achieving an OMPD system. PriD solution i) guarantees confidentiality against inference attacks by defensive mechanism; ii) does not require interaction between users and server by mask matrix; iii) without any approximation by functional encryption. The experimental evaluations illustrate that our PriD scheme can reduce the computation overhead substantially compared with the advanced methods while achieving high precision.
- 8 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet Computing
- Vol. 28 (5) , 47-56
Privacy-Preserving Learning of Prediagnosis Models from Distributed Medical Records
Online Medical Pre-Diagnosis (OMPD) systems based on machine learning will benefit individuals for automatic diagnosis while saving time and costs. The inherent security constraints and distributed setting of patient data form obstacles in constructing reliable OMPD models. Existing privacy-preserving machine learning approaches incur substantial communication and computation cost, and are susceptible to inference attacks. Moreover, to accommodate cryptography primitives, machine learning schemes in the encrypted domain are substituted by approximation models, leading to accuracy loss. To mitigate these problems, we build PriD , an efficient and accurate model for achieving an OMPD system. PriD solution i) guarantees confidentiality against inference attacks by defensive mechanism; ii) does not require interaction between users and server by mask matrix; iii) without any approximation by functional encryption. The experimental evaluations illustrate that our PriD scheme can reduce the computation overhead substantially compared with the advanced methods while achieving high precision.
- 5 April 2023
- Published by Elsevier in Future Generation Computer Systems
- Vol. 145, 454-465
Non-interactive and privacy-preserving neural network learning using functional encryption
- 5 April 2023
- Published by Elsevier in Future Generation Computer Systems
- Vol. 145, 454-465
Non-interactive and privacy-preserving neural network learning using functional encryption
- Open Access
- 14 December 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (24) , 12873
Privacy-Preserving Outsourced Artificial Neural Network Training for Secure Image Classification
Artificial neural network (ANN) is powerful in the artificial intelligence field and has been successfully applied to interpret complex image data in the real world. Since the majority of images are commonly known as private with the information intended to be used by the owner, such as handwritten characters and face, the private constraints form a major obstacle in developing high-precision image classifiers which require access to a large amount of image data belonging to multiple users. State-of-the-art privacy-preserving ANN schemes often use full homomorphic encryption which result in a substantial overhead of computation and data traffic for the data owners, and are restricted to approximation models by low-degree polynomials which lead to a large accuracy loss of the trained model compared to the original ANN model in the plain domain. Consequently, it is still a huge challenge to train an ANN model in the encrypted-domain. To mitigate this problem, we propose a privacy-preserving ANN system for secure constructing image classifiers, named IPPNN, where the server is able to train an ANN-based classifier on the combined image data of all data owners without being able to observe any images using primitives, such as randomization and functional encryption. Our system achieves faster training time and supports lossless training. Moreover, IPPNN removes the need for multiple communications among data owners and servers. We analyze the security of the protocol and perform experiments on a large scale image recognition task. The results show that the IPPNN is feasible to use in practice while achieving high accuracy.
- 14 December 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (24) , 12873
Open AccessPrivacy-Preserving Outsourced Artificial Neural Network Training for Secure Image Classification
Artificial neural network (ANN) is powerful in the artificial intelligence field and has been successfully applied to interpret complex image data in the real world. Since the majority of images are commonly known as private with the information intended to be used by the owner, such as handwritten characters and face, the private constraints form a major obstacle in developing high-precision image classifiers which require access to a large amount of image data belonging to multiple users. State-of-the-art privacy-preserving ANN schemes often use full homomorphic encryption which result in a substantial overhead of computation and data traffic for the data owners, and are restricted to approximation models by low-degree polynomials which lead to a large accuracy loss of the trained model compared to the original ANN model in the plain domain. Consequently, it is still a huge challenge to train an ANN model in the encrypted-domain. To mitigate this problem, we propose a privacy-preserving ANN system for secure constructing image classifiers, named IPPNN, where the server is able to train an ANN-based classifier on the combined image data of all data owners without being able to observe any images using primitives, such as randomization and functional encryption. Our system achieves faster training time and supports lossless training. Moreover, IPPNN removes the need for multiple communications among data owners and servers. We analyze the security of the protocol and perform experiments on a large scale image recognition task. The results show that the IPPNN is feasible to use in practice while achieving high accuracy.