Node selection method in federated learning based on deep reinforcement learning
To cope with the impact of different device computing capabilities and non-independent uniformly distributed data on federated learning performance, and to efficiently schedule terminal devices to complete model aggregation, a method of node selection based on deep reinforcement learning was propose...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2021-06-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2021111/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841539254234120192 |
|---|---|
| author | Wenchen HE Shaoyong GUO Xuesong QIU Liandong CHEN Suxiang ZHANG |
| author_facet | Wenchen HE Shaoyong GUO Xuesong QIU Liandong CHEN Suxiang ZHANG |
| author_sort | Wenchen HE |
| collection | DOAJ |
| description | To cope with the impact of different device computing capabilities and non-independent uniformly distributed data on federated learning performance, and to efficiently schedule terminal devices to complete model aggregation, a method of node selection based on deep reinforcement learning was proposed.It considered training quality and efficiency of heterogeneous terminal devices, and filtrate malicious nodes to guarantee higher model accuracy and shorter training delay of federated learning.Firstly, according to characteristics of model distributed training in federated learning, a node selection system model based on deep reinforcement learning was constructed.Secondly, considering such factors as device training delay, model transmission delay and accuracy, an optimization model of accuracy for node selection was proposed.Finally, the problem model was constructed as a Markov decision process and a node selection algorithm based on distributed proximal strategy optimization was designed to obtain a reasonable set of devices before each training iteration to complete model aggregation.Simulation results demonstrate that the proposed method significantly improves the accuracy and training speed of federated learning, and its convergence and robustness are also well. |
| format | Article |
| id | doaj-art-0a408a717baf4df89e9113d29f82f6f1 |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2021-06-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-0a408a717baf4df89e9113d29f82f6f12025-01-14T07:22:05ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2021-06-0142627159741805Node selection method in federated learning based on deep reinforcement learningWenchen HEShaoyong GUOXuesong QIULiandong CHENSuxiang ZHANGTo cope with the impact of different device computing capabilities and non-independent uniformly distributed data on federated learning performance, and to efficiently schedule terminal devices to complete model aggregation, a method of node selection based on deep reinforcement learning was proposed.It considered training quality and efficiency of heterogeneous terminal devices, and filtrate malicious nodes to guarantee higher model accuracy and shorter training delay of federated learning.Firstly, according to characteristics of model distributed training in federated learning, a node selection system model based on deep reinforcement learning was constructed.Secondly, considering such factors as device training delay, model transmission delay and accuracy, an optimization model of accuracy for node selection was proposed.Finally, the problem model was constructed as a Markov decision process and a node selection algorithm based on distributed proximal strategy optimization was designed to obtain a reasonable set of devices before each training iteration to complete model aggregation.Simulation results demonstrate that the proposed method significantly improves the accuracy and training speed of federated learning, and its convergence and robustness are also well.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2021111/federated learningmodel aggregationnode selectiondeep reinforcement learningaccuracy |
| spellingShingle | Wenchen HE Shaoyong GUO Xuesong QIU Liandong CHEN Suxiang ZHANG Node selection method in federated learning based on deep reinforcement learning Tongxin xuebao federated learning model aggregation node selection deep reinforcement learning accuracy |
| title | Node selection method in federated learning based on deep reinforcement learning |
| title_full | Node selection method in federated learning based on deep reinforcement learning |
| title_fullStr | Node selection method in federated learning based on deep reinforcement learning |
| title_full_unstemmed | Node selection method in federated learning based on deep reinforcement learning |
| title_short | Node selection method in federated learning based on deep reinforcement learning |
| title_sort | node selection method in federated learning based on deep reinforcement learning |
| topic | federated learning model aggregation node selection deep reinforcement learning accuracy |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2021111/ |
| work_keys_str_mv | AT wenchenhe nodeselectionmethodinfederatedlearningbasedondeepreinforcementlearning AT shaoyongguo nodeselectionmethodinfederatedlearningbasedondeepreinforcementlearning AT xuesongqiu nodeselectionmethodinfederatedlearningbasedondeepreinforcementlearning AT liandongchen nodeselectionmethodinfederatedlearningbasedondeepreinforcementlearning AT suxiangzhang nodeselectionmethodinfederatedlearningbasedondeepreinforcementlearning |