… learning paradigm known as Federated Learning (FL), which enables data owners to conduct
model training … device dropouts, the Hierarchical Federated Learning (HFL) framework has …

… In consideration of the dynamics of edge association and resource allocation, we consider a
… Finally in Section VI-C, we study the dynamic resource allocation under varying population …

… client scheduling and resource allocation (ie, the transmit … training data size and thereby
optimize the learning performance of FL. To this end, we propose a dynamic resource allocation …

… stochastic learning algorithms. Although the idea of applying CMDP to … dynamic resource
scheduling is not new, we are motivated to address the resource scheduling issues of resource …

… a novel dynamic resource allocation strategy for energy-efficient federated learning at the …
Hinging on Lyapunov stochastic optimization [24], we develop a dynamic resource allocation …

… Finally, while we consider our approaches primarily in the context of federated learning,
we also demonstrate that q-FFL can be applied to other related problems such as meta-learning, …

… machine learning paradigm called Federated Learning (FL). … device dropouts, the Hierarchical
Federated Learning (HFL) … This decentralized learning approach reduces the reliance …

… can be guaranteed using a learning rate decay, despite the negative effects of the sampling
method. 2) We formulate a resource allocation problem using the proposed FL algorithm in …

… , which significantly improves the learning accuracy. To further improve the performance
of DetFed, we propose a learning-based dynamic resource scheduling algorithm over the …

… resource control problem in the federated learning. To the best our knowledge, this is the first
to translate the federated learning … to cost-effectively accommodate the network dynamics. …