Litcius/Paper detail

Cross-FCL: Toward a Cross-Edge Federated Continual Learning Framework in Mobile Edge Computing Systems

Zhouyangzi Zhang, Bin Guo, Wen Sun, Yan Liu, Zhiwen Yu

2022IEEE Transactions on Mobile Computing27 citationsDOI

Abstract

Federated Learning (FL) in mobile edge computing (MEC) systems has recently been studied extensively. In ubiquitous environments, there are usually cross-edge devices that learn a series of tasks across multiple <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">independent</i> edge FL systems. Due to the differences in the scenarios and tasks of different FL systems, cross-edge devices will forget past tasks after learning new tasks, which is unacceptable for devices that pay system costs to participate in FL. Continual learning (CL) is a viable solution to this problem, which aims to train a model to learn a series of tasks without forgetting old knowledge. Currently, there is no work to investigate the problem of CL in a cross-edge FL scenario. In this paper, we propose <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Cross-FCL</i> , a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Cross</i> -edge <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</i> ederated <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</i> ontinual <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</i> earning framework. Specifically, it enables devices to retain the knowledge learned in the past when participating in new task training through a parameter decomposition based FCL model. Then various cross-edge strategies are introduced, including biased global aggregation and local optimization, to trade off memory and adaptation. We conducted experiments on a real-world dataset and other public datasets. Extensive experiments demonstrate that Cross-FCL achieves best accuracy on IID and highly non-IID tasks with a low storage cost compared to other baselines.

Topics & Concepts

Computer scienceEnhanced Data Rates for GSM EvolutionArtificial intelligencePrivacy-Preserving Technologies in DataDomain Adaptation and Few-Shot LearningMobile Crowdsensing and Crowdsourcing
Cross-FCL: Toward a Cross-Edge Federated Continual Learning Framework in Mobile Edge Computing Systems | Litcius