FedNS: Improving Federated Learning for collaborative image classification on mobile clients
This work addresses the challenge of enhancing collaborative learning efficiency for mobile clients in federated settings, representing an incremental improvement over existing methods.
The paper tackles the problem of improving global model aggregation in federated learning for image classification by proposing FedNS, which filters and re-weights client models at the node/kernel level, resulting in consistent performance gains over FedAvg across multiple datasets and networks.
Federated Learning (FL) is a paradigm that aims to support loosely connected clients in learning a global model collaboratively with the help of a centralized server. The most popular FL algorithm is Federated Averaging (FedAvg), which is based on taking weighted average of the client models, with the weights determined largely based on dataset sizes at the clients. In this paper, we propose a new approach, termed Federated Node Selection (FedNS), for the server's global model aggregation in the FL setting. FedNS filters and re-weights the clients' models at the node/kernel level, hence leading to a potentially better global model by fusing the best components of the clients. Using collaborative image classification as an example, we show with experiments from multiple datasets and networks that FedNS can consistently achieve improved performance over FedAvg.