Ed-Fed: A generic federated learning framework with resource-aware client selection for edge devices

Abstract

Federated learning (FL) has evolved as a prominent method for edge devices to cooperatively create a unified prediction model while securing their sensitive training data local to the device. Despite the existence of numerous research frameworks for simulating FL algorithms, they do not facilitate comprehensive deployment for automatic speech recognition tasks on heterogeneous edge devices. This is where Ed-Fed, a comprehensive and generic FL framework, comes in as a foundation for future practical FL system research. We also propose a novel resource-aware client selection algorithm to optimise the waiting time in the FL settings. We show that our approach can handle the straggler devices and dynamically set the training time for the selected devices in a round. Our evaluation has shown that the proposed approach significantly optimises waiting time in FL compared to conventional random client selection methods.

References in corpus (9)

• TensorFlow: A system for large-scale machine learning
• A Contextual-Bandit Approach to Personalized News Article Recommendation
• Deep Speech 2: End-to-End Speech Recognition in English and Mandarin
• Multi-Armed Bandit Based Client Scheduling for Federated Learning
• TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptation
• Training Speech Recognition Models with Federated Learning: A Quality/Cost Framework
• Neural Contextual Bandits with UCB-based Exploration
• Oort: Efficient Federated Learning via Guided Participant Selection
• Federated Learning in ASR: Not as Easy as You Think