A randomized block policy gradient algorithm with differential privacy in Content Centric Networks
Policy gradient methods are effective means to solve the problems of mobile multimedia data transmission in Content Centric Networks. Current policy gradient algorithms impose high computational cost in processing high-dimensional data. Meanwhile, the issue of privacy disclosure has not been taken i...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-12-01
|
| Series: | International Journal of Distributed Sensor Networks |
| Online Access: | https://doi.org/10.1177/15501477211059934 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832547196560474112 |
|---|---|
| author | Lin Wang Xingang Xu Xuhui Zhao Baozhu Li Ruijuan Zheng Qingtao Wu |
| author_facet | Lin Wang Xingang Xu Xuhui Zhao Baozhu Li Ruijuan Zheng Qingtao Wu |
| author_sort | Lin Wang |
| collection | DOAJ |
| description | Policy gradient methods are effective means to solve the problems of mobile multimedia data transmission in Content Centric Networks. Current policy gradient algorithms impose high computational cost in processing high-dimensional data. Meanwhile, the issue of privacy disclosure has not been taken into account. However, privacy protection is important in data training. Therefore, we propose a randomized block policy gradient algorithm with differential privacy. In order to reduce computational complexity when processing high-dimensional data, we randomly select a block coordinate to update the gradients at each round. To solve the privacy protection problem, we add a differential privacy protection mechanism to the algorithm, and we prove that it preserves the ε -privacy level. We conduct extensive simulations in four environments, which are CartPole, Walker, HalfCheetah, and Hopper. Compared with the methods such as important-sampling momentum-based policy gradient, Hessian-Aided momentum-based policy gradient, REINFORCE, the experimental results of our algorithm show a faster convergence rate than others in the same environment. |
| format | Article |
| id | doaj-art-0622e0a92fe642be8c3319f113f53471 |
| institution | Kabale University |
| issn | 1550-1477 |
| language | English |
| publishDate | 2021-12-01 |
| publisher | Wiley |
| record_format | Article |
| series | International Journal of Distributed Sensor Networks |
| spelling | doaj-art-0622e0a92fe642be8c3319f113f534712025-02-03T06:45:37ZengWileyInternational Journal of Distributed Sensor Networks1550-14772021-12-011710.1177/15501477211059934A randomized block policy gradient algorithm with differential privacy in Content Centric NetworksLin Wang0Xingang Xu1Xuhui Zhao2Baozhu Li3Ruijuan Zheng4Qingtao Wu5School of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaSchool of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaSchool of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaInternet of Things & Smart City Innovation Platform, Zhuhai Fudan Innovation Institute, Zhuhai, ChinaSchool of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaSchool of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaPolicy gradient methods are effective means to solve the problems of mobile multimedia data transmission in Content Centric Networks. Current policy gradient algorithms impose high computational cost in processing high-dimensional data. Meanwhile, the issue of privacy disclosure has not been taken into account. However, privacy protection is important in data training. Therefore, we propose a randomized block policy gradient algorithm with differential privacy. In order to reduce computational complexity when processing high-dimensional data, we randomly select a block coordinate to update the gradients at each round. To solve the privacy protection problem, we add a differential privacy protection mechanism to the algorithm, and we prove that it preserves the ε -privacy level. We conduct extensive simulations in four environments, which are CartPole, Walker, HalfCheetah, and Hopper. Compared with the methods such as important-sampling momentum-based policy gradient, Hessian-Aided momentum-based policy gradient, REINFORCE, the experimental results of our algorithm show a faster convergence rate than others in the same environment.https://doi.org/10.1177/15501477211059934 |
| spellingShingle | Lin Wang Xingang Xu Xuhui Zhao Baozhu Li Ruijuan Zheng Qingtao Wu A randomized block policy gradient algorithm with differential privacy in Content Centric Networks International Journal of Distributed Sensor Networks |
| title | A randomized block policy gradient algorithm with differential privacy in Content Centric Networks |
| title_full | A randomized block policy gradient algorithm with differential privacy in Content Centric Networks |
| title_fullStr | A randomized block policy gradient algorithm with differential privacy in Content Centric Networks |
| title_full_unstemmed | A randomized block policy gradient algorithm with differential privacy in Content Centric Networks |
| title_short | A randomized block policy gradient algorithm with differential privacy in Content Centric Networks |
| title_sort | randomized block policy gradient algorithm with differential privacy in content centric networks |
| url | https://doi.org/10.1177/15501477211059934 |
| work_keys_str_mv | AT linwang arandomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT xingangxu arandomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT xuhuizhao arandomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT baozhuli arandomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT ruijuanzheng arandomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT qingtaowu arandomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT linwang randomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT xingangxu randomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT xuhuizhao randomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT baozhuli randomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT ruijuanzheng randomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks AT qingtaowu randomizedblockpolicygradientalgorithmwithdifferentialprivacyincontentcentricnetworks |