FedGreen: Carbon-aware Federated Learning with Model Size Adaptation

TL;DR

FedGreen is a novel federated learning approach that reduces carbon emissions by adaptively adjusting model sizes based on clients' carbon profiles and locations, balancing environmental impact with model accuracy.

Contribution

This work introduces FedGreen, a carbon-aware federated learning method that employs ordered dropout for adaptive model size sharing based on clients' carbon footprints and locations.

Findings

Significantly reduces carbon footprints compared to existing methods.

Maintains competitive model accuracy despite adaptive model size adjustments.

Provides theoretical analysis of the trade-offs between carbon emissions and convergence accuracy.

Abstract

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.