FedGreen: Carbon-aware Federated Learning with Model Size Adaptation
This work addresses the environmental impact of federated learning for AI practitioners and organizations, though it is incremental as it builds on existing FL and model compression techniques.
The paper tackles the problem of high carbon emissions in federated learning by proposing FedGreen, a method that adapts model sizes based on clients' carbon profiles and locations, which substantially reduces carbon footprints while maintaining competitive model accuracy.
Federated learning (FL) provides a promising collaborative framework to build a model from distributed clients, and this work investigates the carbon emission of the FL process. Cloud and edge servers hosting FL clients may exhibit diverse carbon footprints influenced by their geographical locations with varying power sources, offering opportunities to reduce carbon emissions by training local models with adaptive computations and communications. In this paper, we propose FedGreen, a carbon-aware FL approach to efficiently train models by adopting adaptive model sizes shared with clients based on their carbon profiles and locations using ordered dropout as a model compression technique. We theoretically analyze the trade-offs between the produced carbon emissions and the convergence accuracy, considering the carbon intensity discrepancy across countries to choose the parameters optimally. Empirical studies show that FedGreen can substantially reduce the carbon footprints of FL compared to the state-of-the-art while maintaining competitive model accuracy.