From Eye to Brain: A Proactive and Distributed Crowdsensing Framework for Federated Learning

Abstract

Massive amounts of high-quality data are the prerequisite and support for AI technologies. Due to the nature of privacy-preserving and low communication overheads, federated learning has garnered considerable attention in comparison with traditional data collection methods. However, the performance of federated learning is hampered by the lack of interested clients and limited local data due to selfishness and individual behavioral preferences. To this end, we propose PractFL, a proactive and distributed framework that incorporates the concept of mobile crowdsensing into the federated learning paradigm. Specifically, we design an incentive mechanism in the form of virtual red packets, which are a widely used way of monetary reward and gift-giving in social lives. In this paper we extend this further by giving meaning to the locations, i.e., the red packets are only accessible at specific places. The virtual red packets’ locations and monetary amounts can be dynamically updated by the cloud center to encourage clients to collect additional data that may benefit the federated learning process. Further, we propose a distributed behavioral decision engine based on multi-armed bandits (i.e., choose which red packet to go for) in response to the incentive mechanism enforced by the cloud. Considering the movement cost and conflicts with other clients, K-anonymity and probabilistic selection are introduced in the distributed behavioral decision to recommend the optimal red packet choice for clients without revealing their privacy. The experimental results demonstrate that PractFL outperforms the baselines in terms of classification accuracy. We also find that PractFL can effectively alleviate the overfitting problem caused by class imbalance during the training.