FedKL: Tackling data heterogeneity in federated reinforcement learning by penalizing KL divergence

Z Xie, S Song - IEEE Journal on Selected Areas in …, 2023 - ieeexplore.ieee.org
One of the fundamental issues for Federated Learning (FL) is data heterogeneity, which
causes accuracy degradation, slow convergence, and the communication bottleneck issue …
被引用次数：29 相关文章

FedKL: Tackling Data Heterogeneity in Federated Reinforcement Learning by Penalizing KL Divergence

Z Xie, SH Song - arXiv preprint arXiv:2204.08125, 2022 - arxiv.org
As a distributed learning paradigm, Federated Learning (FL) faces the communication
bottleneck issue due to many rounds of model synchronization and aggregation …

FedKL: Tackling Data Heterogeneity in Federated Reinforcement Learning by Penalizing KL Divergence

Z Xie, SH Song - arXiv e-prints, 2022 - ui.adsabs.harvard.edu
As a distributed learning paradigm, Federated Learning (FL) faces the communication
bottleneck issue due to many rounds of model synchronization and aggregation …

FedKL: Tackling Data Heterogeneity in Federated Reinforcement Learning by Penalizing KL Divergence

Z Xie, S Song - IEEE Journal on Selected Areas in Communications, 2023 - dl.acm.org
One of the fundamental issues for Federated Learning (FL) is data heterogeneity, which
causes accuracy degradation, slow convergence, and the communication bottleneck issue …