Exploration Across Small Silos: Federated Few-Shot Learning on Network Edge

Federated Learning (FL) has been drawing significant attention from both academia and industry working on distributed machine learning. In practice, learning over mutually isolated datasets residing at the network edge, also known as silos, FL clients can suffer a lack of samples, due to many reasons (e.g., expensive annotation), this potentially negative impact performance. Few-Shot (FSL) considered promising solution, but unfortunately cannot be directly applied practical Cross-Silo (CSFL) systems. article, far we know, conduct first systematic discussion specific challenges FSL in CSFL We extract essential design issues found (FFSL), develop new FFSL method based Model-Agnostic Meta (MAML). Through experiments using real-world federated datasets, comprehensively demonstrate our method's advantages existing methods different scenarios where hitherto failed. highlight some future research directions.