A Fair Federated Learning Framework With Reinforcement Learning

doi:doi:10.57702/29fdkcny

A Fair Federated Learning Framework With Reinforcement Learning

Federated learning (FL) is a paradigm where many clients collaboratively train a model under the coordination of a central server, while keeping the training data locally stored. However, heterogeneous data distributions over different clients remain a challenge to mainstream FL algorithms, which may cause slow convergence, overall performance degradation and unfairness of performance across clients.

BibTex:

@dataset{Yaqi_Sun_and_Shijing_Si_and_Jianzong_Wang_and_Yuhan_Dong_and_Zhitao_Zhu_and_Jing_Xiao_2024,
    abstract = {Federated learning (FL) is a paradigm where many clients collaboratively train a model under the coordination of a central server, while keeping the training data locally stored. However, heterogeneous data distributions over different clients remain a challenge to mainstream FL algorithms, which may cause slow convergence, overall performance degradation and unfairness of performance across clients.},
    author = {Yaqi Sun and Shijing Si and Jianzong Wang and Yuhan Dong and Zhitao Zhu and Jing Xiao},
    doi = {10.57702/29fdkcny},
    institution = {No Organization},
    keyword = {'CIFAR-10', 'CIFAR-100', 'Fashion-MNIST'},
    month = {dec},
    publisher = {TIB},
    title = {A Fair Federated Learning Framework With Reinforcement Learning},
    url = {https://service.tib.eu/ldmservice/dataset/a-fair-federated-learning-framework-with-reinforcement-learning},
    year = {2024}
}