RESFL: An Uncertainty-Aware Framework for Responsible Federated Learning by Balancing Privacy, Fairness and Utility in Autonomous Vehicles
Dawood Wasif, Terrence J. Moore, Jin-Hee Cho
TL;DR
RESFL addresses the tension between privacy, fairness, and utility in autonomous-vehicle federated learning by integrating adversarial privacy disentanglement with uncertainty-guided fairness-aware aggregation. By employing an evidential neural network, RESFL quantifies epistemic uncertainty and defines an Uncertainty Fairness Metric (UFM) to weight client updates, prioritizing contributions with lower fairness disparities and higher confidence. The framework leverages a gradient reversal layer to suppress sensitive attribute leakage during training, improving both privacy protection and fairness. Empirical results on FACET and CARLA demonstrate that RESFL achieves competitive detection accuracy while reducing demographic disparities, lowering privacy-attack success rates, and exhibiting greater robustness under adverse weather conditions, making it well-suited for real-world AV deployments.
Abstract
Autonomous vehicles (AVs) increasingly rely on Federated Learning (FL) to enhance perception models while preserving privacy. However, existing FL frameworks struggle to balance privacy, fairness, and robustness, leading to performance disparities across demographic groups. Privacy-preserving techniques like differential privacy mitigate data leakage risks but worsen fairness by restricting access to sensitive attributes needed for bias correction. This work explores the trade-off between privacy and fairness in FL-based object detection for AVs and introduces RESFL, an integrated solution optimizing both. RESFL incorporates adversarial privacy disentanglement and uncertainty-guided fairness-aware aggregation. The adversarial component uses a gradient reversal layer to remove sensitive attributes, reducing privacy risks while maintaining fairness. The uncertainty-aware aggregation employs an evidential neural network to weight client updates adaptively, prioritizing contributions with lower fairness disparities and higher confidence. This ensures robust and equitable FL model updates. We evaluate RESFL on the FACET dataset and CARLA simulator, assessing accuracy, fairness, privacy resilience, and robustness under varying conditions. RESFL improves detection accuracy, reduces fairness disparities, and lowers privacy attack success rates while demonstrating superior robustness to adversarial conditions compared to other approaches.