Mitigating Group Bias in Federated Learning: Beyond Local Fairness
This work addresses group bias in federated learning, which is a critical issue for ensuring fairness in distributed machine learning systems, though it appears incremental by building on existing local fairness approaches.
The paper tackled the problem of group fairness in federated learning by investigating the limitations of local fairness methods and proposing a globally fair training algorithm that directly minimizes penalized empirical loss. The result showed promising performance in enhancing fairness while retaining high accuracy compared to locally fair methods, as demonstrated in real-data experiments.
The issue of group fairness in machine learning models, where certain sub-populations or groups are favored over others, has been recognized for some time. While many mitigation strategies have been proposed in centralized learning, many of these methods are not directly applicable in federated learning, where data is privately stored on multiple clients. To address this, many proposals try to mitigate bias at the level of clients before aggregation, which we call locally fair training. However, the effectiveness of these approaches is not well understood. In this work, we investigate the theoretical foundation of locally fair training by studying the relationship between global model fairness and local model fairness. Additionally, we prove that for a broad class of fairness metrics, the global model's fairness can be obtained using only summary statistics from local clients. Based on that, we propose a globally fair training algorithm that directly minimizes the penalized empirical loss. Real-data experiments demonstrate the promising performance of our proposed approach for enhancing fairness while retaining high accuracy compared to locally fair training methods.