Differential privacy budget optimization based on deep learning in IoT
In order to effectively process the massive data brought by the large-scale application of the internet of things (IoT), deep learning is widely used in IoT environment.However, in the training process of deep learning, there are security threats such as reasoning attacks and model reverse attacks,...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
China InfoCom Media Group
2022-06-01
|
| Series: | 物联网学报 |
| Subjects: | |
| Online Access: | http://www.wlwxb.com.cn/zh/article/doi/10.11959/j.issn.2096-3750.2022.00264/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841531180914049024 |
|---|---|
| author | Dan LUO Ruzhi XU Zhitao GUAN |
| author_facet | Dan LUO Ruzhi XU Zhitao GUAN |
| author_sort | Dan LUO |
| collection | DOAJ |
| description | In order to effectively process the massive data brought by the large-scale application of the internet of things (IoT), deep learning is widely used in IoT environment.However, in the training process of deep learning, there are security threats such as reasoning attacks and model reverse attacks, which can lead to the leakage of the original data input to the model.Applying differential privacy to protect the training process parameters of the deep model is an effective way to solve this problem.A differential privacy budget optimization method was proposed based on deep learning in IoT, which adaptively allocates different budgets according to the iterative change of parameters.In order to avoid the excessive noise, a regularization term was introduced to constrain the disturbance term.Preventing the neural network from over fitting also helps to learn the salient features of the model.Experiments show that this method can effectively enhance the generalization ability of the model.As the number of iterations increases, the accuracy of the model trained after adding noise is almost the same as that obtained by training using the original data, which not only achieves privacy protection, but also guarantees the availability, which means balance the privacy and availability. |
| format | Article |
| id | doaj-art-8ce2de1bc2f84c9ca9c97cac305e3ead |
| institution | Kabale University |
| issn | 2096-3750 |
| language | zho |
| publishDate | 2022-06-01 |
| publisher | China InfoCom Media Group |
| record_format | Article |
| series | 物联网学报 |
| spelling | doaj-art-8ce2de1bc2f84c9ca9c97cac305e3ead2025-01-15T02:53:24ZzhoChina InfoCom Media Group物联网学报2096-37502022-06-016657659648540Differential privacy budget optimization based on deep learning in IoTDan LUORuzhi XUZhitao GUANIn order to effectively process the massive data brought by the large-scale application of the internet of things (IoT), deep learning is widely used in IoT environment.However, in the training process of deep learning, there are security threats such as reasoning attacks and model reverse attacks, which can lead to the leakage of the original data input to the model.Applying differential privacy to protect the training process parameters of the deep model is an effective way to solve this problem.A differential privacy budget optimization method was proposed based on deep learning in IoT, which adaptively allocates different budgets according to the iterative change of parameters.In order to avoid the excessive noise, a regularization term was introduced to constrain the disturbance term.Preventing the neural network from over fitting also helps to learn the salient features of the model.Experiments show that this method can effectively enhance the generalization ability of the model.As the number of iterations increases, the accuracy of the model trained after adding noise is almost the same as that obtained by training using the original data, which not only achieves privacy protection, but also guarantees the availability, which means balance the privacy and availability.http://www.wlwxb.com.cn/zh/article/doi/10.11959/j.issn.2096-3750.2022.00264/IoTdifferential privacyregularizationdeep learningprivacy budget |
| spellingShingle | Dan LUO Ruzhi XU Zhitao GUAN Differential privacy budget optimization based on deep learning in IoT 物联网学报 IoT differential privacy regularization deep learning privacy budget |
| title | Differential privacy budget optimization based on deep learning in IoT |
| title_full | Differential privacy budget optimization based on deep learning in IoT |
| title_fullStr | Differential privacy budget optimization based on deep learning in IoT |
| title_full_unstemmed | Differential privacy budget optimization based on deep learning in IoT |
| title_short | Differential privacy budget optimization based on deep learning in IoT |
| title_sort | differential privacy budget optimization based on deep learning in iot |
| topic | IoT differential privacy regularization deep learning privacy budget |
| url | http://www.wlwxb.com.cn/zh/article/doi/10.11959/j.issn.2096-3750.2022.00264/ |
| work_keys_str_mv | AT danluo differentialprivacybudgetoptimizationbasedondeeplearninginiot AT ruzhixu differentialprivacybudgetoptimizationbasedondeeplearninginiot AT zhitaoguan differentialprivacybudgetoptimizationbasedondeeplearninginiot |