FedLD

Implementation of the FedLD [FedLD24] algorithm.

References

[FedLD24]

Shuang Zeng, Pengxin Guo, Shuai Wang, Jianbo Wang, Yuyin Zhou and Liangqiong Qu. Tackling Data Heterogeneity in Federated Learning via Loss Decomposition. In MICCAI (2024). URL: https://papers.miccai.org/miccai-2024/paper/1348_paper.pdf

Classes included in fluke.algorithms.fedld

FedLDClient

FedLDServer

FedLD

Classes

class fluke.algorithms.fedld.FedLDClient(index: int, train_set: FastDataLoader, test_set: FastDataLoader, optimizer_cfg: OptimizerConfigurator, loss_fn: Module, local_epochs: int = 3, fine_tuning_epochs: int = 0, clipping: float = 0, margin_lam: float = 0.2, **kwargs: dict[str, Any])[source]

Bases: Client

class fluke.algorithms.fedld.FedLDServer(model: Module, test_set: FastDataLoader, clients: Iterable[Client], weighted: bool = False, lr: float = 1.0, k_proportion: float = 0.8, **kwargs: dict[str, Any])[source]

Bases: Server

class fluke.algorithms.fedld.FedLD(n_clients: int, data_splitter: DataSplitter, hyper_params: DDict | dict[str, Any], **kwargs: dict[str, Any])[source]

Bases: CentralizedFL