FLARE: Detection and Mitigation of Concept Drift for Federated Learning based IoT Deployments
This addresses the challenge of maintaining model performance in dynamic IoT environments while reducing communication overhead, though it appears incremental as it builds on existing federated learning and concept drift detection approaches.
The paper tackles the problem of concept drift in federated learning for IoT deployments by introducing FLARE, a lightweight dual-scheduler framework that conditionally transfers data and deploys models based on training behavior and inference statistics. The results show over 5x reduction in data exchange compared to fixed-interval methods and at least 16x reduction in latency for concept drift detection.
Intelligent, large-scale IoT ecosystems have become possible due to recent advancements in sensing technologies, distributed learning, and low-power inference in embedded devices. In traditional cloud-centric approaches, raw data is transmitted to a central server for training and inference purposes. On the other hand, Federated Learning migrates both tasks closer to the edge nodes and endpoints. This allows for a significant reduction in data exchange while preserving the privacy of users. Trained models, though, may under-perform in dynamic environments due to changes in the data distribution, affecting the model's ability to infer accurately; this is referred to as concept drift. Such drift may also be adversarial in nature. Therefore, it is of paramount importance to detect such behaviours promptly. In order to simultaneously reduce communication traffic and maintain the integrity of inference models, we introduce FLARE, a novel lightweight dual-scheduler FL framework that conditionally transfers training data, and deploys models between edge and sensor endpoints based on observing the model's training behaviour and inference statistics, respectively. We show that FLARE can significantly reduce the amount of data exchanged between edge and sensor nodes compared to fixed-interval scheduling methods (over 5x reduction), is easily scalable to larger systems, and can successfully detect concept drift reactively with at least a 16x reduction in latency.