Secure federated learning scheme based on adaptive Byzantine defense
Aiming at the problem that the existing federated learning schemes cannot adaptively defend Byzantine attacks and low model accuracy, a secure federated learning scheme based on adaptive Byzantine defense was proposed. Through adaptive preliminary aggregation associated with incentives and global ag...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2024-08-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024138/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841539168405028864 |
|---|---|
| author | ZHOU Yousheng GAO Jingkun ZUO Xiangjian LIU Yuanni |
| author_facet | ZHOU Yousheng GAO Jingkun ZUO Xiangjian LIU Yuanni |
| author_sort | ZHOU Yousheng |
| collection | DOAJ |
| description | Aiming at the problem that the existing federated learning schemes cannot adaptively defend Byzantine attacks and low model accuracy, a secure federated learning scheme based on adaptive Byzantine defense was proposed. Through adaptive preliminary aggregation associated with incentives and global aggregation based on exponential weighted average, the global model was minimally perturbed on the premise of providing differential privacy perturbations for both the local model and the global model to achieve privacy protection. Different penalties were given to Byzantine client local models to adaptively defend Byzantine attacks, mobilized the enthusiasm of participants, and achieved higher model accuracy. Experimental results show that for different proportions of Byzantine clients, comparing the proposed scheme with other comparative schemes, the model accuracy is increased by 3.51%, 3.55% and 5.12% on average respectively, achieving higher model accuracy when adaptively defending Byzantine attacks. |
| format | Article |
| id | doaj-art-0b83207b7e4f42058ef720204822c562 |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2024-08-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-0b83207b7e4f42058ef720204822c5622025-01-14T07:24:49ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2024-08-014516617969425437Secure federated learning scheme based on adaptive Byzantine defenseZHOU YoushengGAO JingkunZUO XiangjianLIU YuanniAiming at the problem that the existing federated learning schemes cannot adaptively defend Byzantine attacks and low model accuracy, a secure federated learning scheme based on adaptive Byzantine defense was proposed. Through adaptive preliminary aggregation associated with incentives and global aggregation based on exponential weighted average, the global model was minimally perturbed on the premise of providing differential privacy perturbations for both the local model and the global model to achieve privacy protection. Different penalties were given to Byzantine client local models to adaptively defend Byzantine attacks, mobilized the enthusiasm of participants, and achieved higher model accuracy. Experimental results show that for different proportions of Byzantine clients, comparing the proposed scheme with other comparative schemes, the model accuracy is increased by 3.51%, 3.55% and 5.12% on average respectively, achieving higher model accuracy when adaptively defending Byzantine attacks.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024138/federated learningedge computingsecurity and privacy protectionByzantine attack |
| spellingShingle | ZHOU Yousheng GAO Jingkun ZUO Xiangjian LIU Yuanni Secure federated learning scheme based on adaptive Byzantine defense Tongxin xuebao federated learning edge computing security and privacy protection Byzantine attack |
| title | Secure federated learning scheme based on adaptive Byzantine defense |
| title_full | Secure federated learning scheme based on adaptive Byzantine defense |
| title_fullStr | Secure federated learning scheme based on adaptive Byzantine defense |
| title_full_unstemmed | Secure federated learning scheme based on adaptive Byzantine defense |
| title_short | Secure federated learning scheme based on adaptive Byzantine defense |
| title_sort | secure federated learning scheme based on adaptive byzantine defense |
| topic | federated learning edge computing security and privacy protection Byzantine attack |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024138/ |
| work_keys_str_mv | AT zhouyousheng securefederatedlearningschemebasedonadaptivebyzantinedefense AT gaojingkun securefederatedlearningschemebasedonadaptivebyzantinedefense AT zuoxiangjian securefederatedlearningschemebasedonadaptivebyzantinedefense AT liuyuanni securefederatedlearningschemebasedonadaptivebyzantinedefense |