Polaris: Accelerating Asynchronous Federated Learning With Client Selection

Federated learning has garnered significant research attention as a privacy-preserving learning paradigm. Asynchronous federated learning has been proposed to improve scalability by accommodating slower clients, commonly referred to as stragglers. However, asynchronous federated learning suffers from slow convergence with respect to wall-clock time, due to the existence of data heterogeneity and staleness. Existing strategies struggled to tackle both difficulties for a wide range of deep learning models. To address the problem, we propose
Polaris
, a theoretically sound design and a new take to client selection for asynchronous federated learning. With
Polaris
, we first theoretically investigated the design space of client sampling strategies from a geometric optimization perspective, taking both data heterogeneity and staleness into account. Our design is not only theoretically proven, but also thoroughly tested in our reproducible experimental open-source testbed. Our experimental results demonstrates overwhelming evidence that
Polaris
outperformed existing state-of-the-art client selection strategies by a substantial margin over a wide variety of tasks and datasets, as we train image classification models using
CIFAR-10
,
CIFAR-100
,
CINIC-10
,
Federated EMNIST
, and a language modeling model using the
Tiny Shakespeare
dataset. Further, our extensive array of ablation studies have also shown that
Polaris
is both scalable and robust as the size of datasets scale up and data heterogeneity vary.