Privacy protected federated learning improves vehicle image recognition
Privacy-Preserving Federated Loss Exploration for Vision-Based Perception under Non-IID Data
Cryptography and Security
Summary
Self-driving cars use many cameras to recognize traffic signs, but sharing all data to improve recognition risks revealing private information. The authors found some existing learning methods leak sensitive details when sharing data hints to improve accuracy. They created a new method called PPFLex that uses multiple secure servers to combine learning safely without exposing private information. Their method matches the accuracy of existing approaches while preventing data leaks and only adds a manageable amount of extra processing.
What this means in practice
- •For autonomous vehicle engineers: Train traffic-sign classifiers from multiple vehicles without risking client privacy leakage by adopting PPFLex's secure federated learning approach.
- •For smart city operators: Combine camera data from dispersed roadside sensors securely for traffic monitoring without exposing sensitive location-specific data using multi-party computation methods.
Authors
Nergiz Yuca, Christian Internò, Nikolay Matyunin, Markus Olhofer, Barbara Hammer, Stefan Katzenbeisser
Abstract
Vision-based image classification models benefit from Federated Learning (FL) across vehicles, however learning accurate models from distributed vehicular cameras is challenging due to data heterogeneity and privacy concerns. To improve performance on non-IID data, advanced FL optimization approaches have emerged that share auxiliary insights, such as client gradient behaviors. While these gradient-guided optimized FL methods improve the model accuracy, their reliance on sharing additional gradient information introduces a previously unexplored privacy vulnerability. In our empirical privacy-leakage experiment, we show that a malicious server can infer clients' label sets from shared guidance matrices under heterogeneous client label distributions (simulated via Dirichlet-based label partitions), between 50% and 84% accuracy. To address this privacy leakage, we propose PPFLex that replaces the central server with three Secure Multi-Party Computation (MPC) servers to securely compute the global guidance matrix and perform federated averaging under semi-honest and malicious MPC server assumptions. We evaluate PPFLex on widely used traffic-sign classification datasets (e.g., GTSRB and BTSC) as representative autonomous vehicle perception benchmarks, comparing against baselines. Our results show that PPFLex achieves accuracy comparable to the plain-text FedLEx pipeline while preventing privacy leakage, incurring acceptable MPC overhead.