Crowdsourced federated learning architecture with personalized privacy preservation

In crowdsourced federated learning, differential privacy is commonly used to prevent the aggregation server from recovering training data from the models uploaded by clients to achieve privacy preservation. However, improper privacy budget settings and perturbation methods will severely impact model...

Deskribapen osoa

Xehetasun bibliografikoak

Egile Nagusiak:	Yunfan Xu, Xuesong Qiu, Fan Zhang, Jiakai Hao
Formatua:	Artigo
Hizkuntza:	Inglês
Argitaratua:	Tsinghua University Press 2024-09-01
Saila:	Intelligent and Converged Networks
Gaiak:	crowdsourced federated learning differential privacy privacy preservation two-stage game
Sarrera elektronikoa:	https://www.sciopen.com/article/10.23919/ICN.2024.0014
Etiketak:	Etiketa erantsi Etiketarik gabe, Izan zaitez lehena erregistro honi etiketa jartzen!