An adaptive client selection method for long tailed scene classification in autonomous driving
- 1School of Artificial Intelligence, HuBei Open University
- 2School of Computer Engineering, HuBei University of Arts and Science
Abstract
Collaborative autonomous driving utilizing Federated Learning (FL) is frequently constrained by long-tailed data distributions and perceptual imbalances. Implementing federated learning for environmental scene classification in autonomous driving faces severe challenges due to the long-tailed and non-IID distribution of real-world climatic data. To address this, this paper proposes FedRare, a client selection framework based on a multi-dimensional utility function, integrating scenario criticality ( WC ), distribution rarity ( WR ), and local update quality ( WL ). To address the safety concerns in edge cases, we implement a granular evaluation protocol by categorizing 18 driving scenarios into standard, diverse, and safety-critical groups. Experiments on the BDD100K dataset show that FedRare achieves a mean recall of 72.82% during the late training stage (averaged over the final five communication rounds across five random seeds) while the baseline is 63.42%. Results from the trained model validation demonstrate that FedRare achieves a 74.13% recall in safety-critical groups (e.g., rainy/snowy night), while maintaining high perceptual reliability where traditional loss-based methods often fail. Furthermore, the WL constraint stabilizes the training process, keeping the performance fluctuation (standard deviation) at approximately 0.02. This work proposes a framework to enhance perceptual robustness for autonomous driving within complex, long-tailed environments.
Related
- Gallstone Status Classification Using Adaptive Performance-Diversity Hybrid Ensemble with Pareto-Based Multi-Objective Selection (APDHE-Pareto)
- Noise Conditioned Bayesian Optimization Strategy Selection for Self-Driving Labs
- An adaptive cruise control system for autonomous vehicles using PRDC–CoROA-based hybrid optimization
- Adaptive self-supervised learning for real-time problem solving in autonomous systems
- Adaptive Liquid State-Space Networks: A Novel Neuro-Dynamic Architecture for Tabular Classification with Application to Anemia Diagnosis
BibTeX
@article{wen2026adaptive,
title = {{An adaptive client selection method for long tailed scene classification in autonomous driving}},
author = {Song Wen and Xinying Wang},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62562-8},
url = {https://doi.org/10.1038/s41598-026-62562-8},
note = {CC BY},
}
RIS
TY - JOUR
TI - An adaptive client selection method for long tailed scene classification in autonomous driving
AU - Song Wen
AU - Xinying Wang
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62562-8
UR - https://doi.org/10.1038/s41598-026-62562-8
AB - Collaborative autonomous driving utilizing Federated Learning (FL) is frequently constrained by long-tailed data distributions and perceptual imbalances. Implementing federated learning for environmental scene classification in autonomous driving faces severe challenges due to the long-tailed and non-IID distribution of real-world climatic data. To address this, this paper proposes FedRare, a client selection framework based on a multi-dimensional utility function, integrating scenario criticality ( W_C ), distribution rarity ( W_R ), and local update quality ( W_L ). To address the safety concerns in edge cases, we implement a granular evaluation protocol by categorizing 18 driving scenarios into standard, diverse, and safety-critical groups. Experiments on the BDD100K dataset show that FedRare achieves a mean recall of 72.82% during the late training stage (averaged over the final five communication rounds across five random seeds) while the baseline is 63.42%. Results from the trained model validation demonstrate that FedRare achieves a 74.13% recall in safety-critical groups (e.g., rainy/snowy night), while maintaining high perceptual reliability where traditional loss-based methods often fail. Furthermore, the W_L constraint stabilizes the training process, keeping the performance fluctuation (standard deviation) at approximately 0.02. This work proposes a framework to enhance perceptual robustness for autonomous driving within complex, long-tailed environments.
SN - 2045-2322
ER -