1: \begin{abstract}

2: Federated Learning (FL) is an emerging machine learning technique that enables distributed model training across data silos or edge devices without data sharing.

3: Yet, FL inevitably introduces inefficiencies compared to centralized model training, which will further increase the already high energy usage and associated carbon emissions of machine learning in the future.

4: Although the scheduling of workloads based on the availability of low-carbon energy has received considerable attention in recent years, it has not yet been investigated in the context of FL.

5: However, FL is a highly promising use case for carbon-aware computing, as training jobs constitute of energy-intensive batch processes scheduled in geo-distributed environments.

6: 

7: We propose FedZero, a FL system that operates exclusively on renewable excess energy and spare capacity of compute infrastructure to effectively reduce the training's operational carbon emissions to zero.

8: Based on energy and load forecasts, FedZero leverages the spatio-temporal availability of excess energy by cherry-picking clients for fast convergence and fair participation.

9: Our evaluation, based on real solar and load traces, shows that FedZero converges considerably faster under the mentioned constraints than state-of-the-art approaches, is highly scalable, and is robust against forecasting errors.

10: \end{abstract}

11: