A Symmetric Projection Space and Adversarial Training Framework for Privacy-Preserving Machine Learning with Improved Computational Efficiency
This paper proposes a data security training framework based on symmetric projection space and adversarial training, aimed at addressing the issues of privacy leakage and computational efficiency encountered by current privacy protection technologies when processing sensitive data. By designing a ne...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-03-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/6/3275 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849341806673657856 |
|---|---|
| author | Qianqian Li Shutian Zhou Xiangrong Zeng Jiaqi Shi Qianye Lin Chenjia Huang Yuchen Yue Yuyao Jiang Chunli Lv |
| author_facet | Qianqian Li Shutian Zhou Xiangrong Zeng Jiaqi Shi Qianye Lin Chenjia Huang Yuchen Yue Yuyao Jiang Chunli Lv |
| author_sort | Qianqian Li |
| collection | DOAJ |
| description | This paper proposes a data security training framework based on symmetric projection space and adversarial training, aimed at addressing the issues of privacy leakage and computational efficiency encountered by current privacy protection technologies when processing sensitive data. By designing a new projection loss function and combining autoencoders with adversarial training, the proposed method effectively balances privacy protection and model utility. Experimental results show that, for financial time-series data tasks, the model using the projection loss achieves a precision of 0.95, recall of 0.91, and accuracy of 0.93, significantly outperforming the traditional cross-entropy loss. In image data tasks, the projection loss yields a precision of 0.93, recall of 0.90, accuracy of 0.91, and mAP@50 and mAP@75 of 0.91 and 0.90, respectively, demonstrating its strong advantage in complex tasks. Furthermore, experiments on different hardware platforms (Raspberry Pi, Jetson, and NVIDIA 3080 GPU) show that the proposed method performs well on low-computation devices and exhibits significant advantages on high-performance GPUs, particularly in terms of computational efficiency, demonstrating good scalability and efficiency. The experimental results validate the superiority of the proposed method in terms of data privacy protection and computational efficiency. |
| format | Article |
| id | doaj-art-41280c3ba6c14e56b2e096eac6887daf |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-41280c3ba6c14e56b2e096eac6887daf2025-08-20T03:43:33ZengMDPI AGApplied Sciences2076-34172025-03-01156327510.3390/app15063275A Symmetric Projection Space and Adversarial Training Framework for Privacy-Preserving Machine Learning with Improved Computational EfficiencyQianqian Li0Shutian Zhou1Xiangrong Zeng2Jiaqi Shi3Qianye Lin4Chenjia Huang5Yuchen Yue6Yuyao Jiang7Chunli Lv8China Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaThis paper proposes a data security training framework based on symmetric projection space and adversarial training, aimed at addressing the issues of privacy leakage and computational efficiency encountered by current privacy protection technologies when processing sensitive data. By designing a new projection loss function and combining autoencoders with adversarial training, the proposed method effectively balances privacy protection and model utility. Experimental results show that, for financial time-series data tasks, the model using the projection loss achieves a precision of 0.95, recall of 0.91, and accuracy of 0.93, significantly outperforming the traditional cross-entropy loss. In image data tasks, the projection loss yields a precision of 0.93, recall of 0.90, accuracy of 0.91, and mAP@50 and mAP@75 of 0.91 and 0.90, respectively, demonstrating its strong advantage in complex tasks. Furthermore, experiments on different hardware platforms (Raspberry Pi, Jetson, and NVIDIA 3080 GPU) show that the proposed method performs well on low-computation devices and exhibits significant advantages on high-performance GPUs, particularly in terms of computational efficiency, demonstrating good scalability and efficiency. The experimental results validate the superiority of the proposed method in terms of data privacy protection and computational efficiency.https://www.mdpi.com/2076-3417/15/6/3275privacy protectiondata securityadversarial trainingcomputational efficiencyhigh-dimensional data compression |
| spellingShingle | Qianqian Li Shutian Zhou Xiangrong Zeng Jiaqi Shi Qianye Lin Chenjia Huang Yuchen Yue Yuyao Jiang Chunli Lv A Symmetric Projection Space and Adversarial Training Framework for Privacy-Preserving Machine Learning with Improved Computational Efficiency Applied Sciences privacy protection data security adversarial training computational efficiency high-dimensional data compression |
| title | A Symmetric Projection Space and Adversarial Training Framework for Privacy-Preserving Machine Learning with Improved Computational Efficiency |
| title_full | A Symmetric Projection Space and Adversarial Training Framework for Privacy-Preserving Machine Learning with Improved Computational Efficiency |
| title_fullStr | A Symmetric Projection Space and Adversarial Training Framework for Privacy-Preserving Machine Learning with Improved Computational Efficiency |
| title_full_unstemmed | A Symmetric Projection Space and Adversarial Training Framework for Privacy-Preserving Machine Learning with Improved Computational Efficiency |
| title_short | A Symmetric Projection Space and Adversarial Training Framework for Privacy-Preserving Machine Learning with Improved Computational Efficiency |
| title_sort | symmetric projection space and adversarial training framework for privacy preserving machine learning with improved computational efficiency |
| topic | privacy protection data security adversarial training computational efficiency high-dimensional data compression |
| url | https://www.mdpi.com/2076-3417/15/6/3275 |
| work_keys_str_mv | AT qianqianli asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT shutianzhou asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT xiangrongzeng asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT jiaqishi asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT qianyelin asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT chenjiahuang asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT yuchenyue asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT yuyaojiang asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT chunlilv asymmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT qianqianli symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT shutianzhou symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT xiangrongzeng symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT jiaqishi symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT qianyelin symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT chenjiahuang symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT yuchenyue symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT yuyaojiang symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency AT chunlilv symmetricprojectionspaceandadversarialtrainingframeworkforprivacypreservingmachinelearningwithimprovedcomputationalefficiency |