FAIR-FATE: Fair Federated Learning with Momentum
This addresses fairness issues for unprivileged groups in decentralized machine learning settings, representing an incremental advance by applying a momentum-based approach to a known bottleneck in federated learning.
The authors tackled the problem of group fairness in federated learning with heterogeneous client data by proposing FAIR-FATE, a novel algorithm that uses a fairness-aware aggregation method with a momentum term to estimate fair model updates, resulting in outperforming state-of-the-art fair federated learning algorithms on real-world datasets.
While fairness-aware machine learning algorithms have been receiving increasing attention, the focus has been on centralized machine learning, leaving decentralized methods underexplored. Federated Learning is a decentralized form of machine learning where clients train local models with a server aggregating them to obtain a shared global model. Data heterogeneity amongst clients is a common characteristic of Federated Learning, which may induce or exacerbate discrimination of unprivileged groups defined by sensitive attributes such as race or gender. In this work we propose FAIR-FATE: a novel FAIR FederATEd Learning algorithm that aims to achieve group fairness while maintaining high utility via a fairness-aware aggregation method that computes the global model by taking into account the fairness of the clients. To achieve that, the global model update is computed by estimating a fair model update using a Momentum term that helps to overcome the oscillations of non-fair gradients. To the best of our knowledge, this is the first approach in machine learning that aims to achieve fairness using a fair Momentum estimate. Experimental results on real-world datasets demonstrate that FAIR-FATE outperforms state-of-the-art fair Federated Learning algorithms under different levels of data heterogeneity.