发表机构
Yonsei University; Toyota Motor Corporation（延世大学; 丰田汽车公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对车载联邦学习中车辆离开导致类别样本稀缺、识别性能下降的问题，提出MPT框架，通过重心分解、协方差残差预测和自适应校准，从隐私保护的类别级统计量重构原型，在nuImages数据集上以1%稀有样本达到0.516的F1分数。
AI 中文摘要
跨车辆联邦学习使车辆能够在保持本地采集的驾驶数据私密性的同时，协作改进感知模型。然而，车辆参与是短暂的，一辆车可能在训练收敛之前离开，并永久带走其本地数据。当这辆离开的车辆持有目标类别的大多数样本时，该类别在剩余的联邦学习网络中变得稀有，并且随着共享骨干网络的持续演化，其识别性能可能悄然下降。由于剩余的少量样本提供了有噪声的原型估计，而联邦学习的隐私约束阻止了集中访问原始数据或逐样本特征，因此恢复该类别是困难的。本文提出了MPT，一种跨车辆联邦学习框架，通过每轮从保护隐私的类别级统计量中重构稀有类别的原型来维持其识别能力。MPT结合了重心分解（用于跟踪与剩余类别原型共享的漂移）、基于协方差的残差预测（用于估计超出范围的漂移）以及自适应校准（根据可靠性对剩余稀有类别样本进行加权）。我们在三个车辆分类任务和四个骨干网络上，与代表性的校准和漂移补偿基线进行了评估。MPT在稀有类别F1分数上优于所有基线，在nuImages数据集上仅保留1%的稀有类别样本时达到0.516，且无需原始数据、逐样本特征或重新训练。
英文摘要
Cross-vehicle federated learning enables vehicles to collaboratively improve perception models while keeping locally collected driving data private. However, vehicle participation is transient, and a vehicle may depart before training converges while permanently taking its local data. When this departing vehicle holds most samples of a target class, the class becomes rare in the remaining FL network, and its recognition can silently degrade as the shared backbone continues to evolve. Recovering the class is difficult since the few remaining samples provide a noisy prototype estimate, while FL privacy constraints prevent centralized access to raw data or per-sample features. This paper presents MPT, a cross-vehicle FL framework that maintains rare-class recognition by reconstructing its prototype at every round from privacy-preserving class-level statistics. MPT combines a barycentric decomposition that tracks drift shared with remaining-class prototypes, a covariance-based residual prediction that estimates out-of-span drift, and an adaptive calibration that weighs the remaining rare-class samples according to their reliability. We evaluate MPT on three vehicle classification tasks and four backbones against representative calibration and drift-compensation baselines. MPT outperforms all baselines in rare class F1, reaching 0.516 on the nuImages dataset with only 1\% of rare-class samples remaining, without raw data, per-sample features, or retraining.
Comments9 pages, 7 figures
