FedDSR: Federated Deep Supervision and Regularization Towards Autonomous Driving

2025-12-07

AI Summary

To address poor generalization and slow convergence caused by non-independent and identically distributed (Non-IID) data in autonomous driving federated learning, this paper proposes FedDSR—a novel framework that jointly leverages mutual information estimation and negentropy regularization over multi-layer intermediate features. FedDSR establishes a hierarchical intermediate supervision and collaborative regularization mechanism to optimize feature representations layer-wise. It is compatible with mainstream federated algorithms and network architectures, and applicable to tasks such as semantic segmentation. Experiments on multiple Non-IID autonomous driving datasets demonstrate that FedDSR achieves up to an 8.93% improvement in mean Intersection-over-Union (mIoU) over baseline methods while reducing communication rounds by 28.57%. These results confirm its effectiveness in significantly enhancing model generalization and training efficiency.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Federated Learning (FL) enables collaborative training of autonomous driving (AD) models across distributed vehicles while preserving data privacy. However, FL encounters critical challenges such as poor generalization and slow convergence due to non-independent and identically distributed (non-IID) data from diverse driving environments. To overcome these obstacles, we introduce Federated Deep Supervision and Regularization (FedDSR), a paradigm that incorporates multi-access intermediate layer supervision and regularization within federated AD system. Specifically, FedDSR comprises following integral strategies: (I) to select multiple intermediate layers based on predefined architecture-agnostic standards. (II) to compute mutual information (MI) and negative entropy (NE) on those selected layers to serve as intermediate loss and regularizer. These terms are integrated into the output-layer loss to form a unified optimization objective, enabling comprehensive optimization across the network hierarchy. (III) to aggregate models from vehicles trained based on aforementioned rules of (I) and (II) to generate the global model on central server. By guiding and penalizing the learning of feature representations at intermediate stages, FedDSR enhances the model generalization and accelerates model convergence for federated AD. We then take the semantic segmentation task as an example to assess FedDSR and apply FedDSR to multiple model architectures and FL algorithms. Extensive experiments demonstrate that FedDSR achieves up to 8.93% improvement in mIoU and 28.57% reduction in training rounds, compared to other FL baselines, making it highly suitable for practical deployment in federated AD ecosystems.

Problem

• Addresses poor generalization in federated autonomous driving due to non-IID data
• Accelerates convergence in federated learning for autonomous driving systems
• Enhances model performance via intermediate layer supervision and regularization

Innovation

• Multi-layer supervision and regularization in federated learning
• Mutual information and negative entropy for intermediate loss
• Enhanced generalization and convergence for autonomous driving