FedCLF - Towards Efficient Participant Selection for Federated Learning in Heterogeneous IoV Networks

2025-09-25

AI Summary

In vehicular networks (IoV), the dual heterogeneity of data and devices severely degrades model accuracy and resource utilization in federated learning (FL). To address this, we propose a dynamic participant selection mechanism based on calibrated loss and feedback control. Specifically, we introduce calibrated loss as a novel metric to quantify client utility and design a feedback controller to dynamically adjust client sampling frequency—thereby jointly optimizing communication and computation overhead while guaranteeing model convergence. Extensive experiments under high-data-heterogeneity settings using CIFAR-10 demonstrate that our method achieves up to a 16% improvement in model accuracy over FedAvg, Newton-based FL (Newt), and Oort, while significantly reducing client sampling frequency. This work thus bridges the gap between statistical performance and system efficiency in resource-constrained IoV environments.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Federated Learning (FL) is a distributed machine learning technique that preserves data privacy by sharing only the trained parameters instead of the client data. This makes FL ideal for highly dynamic, heterogeneous, and time-critical applications, in particular, the Internet of Vehicles (IoV) networks. However, FL encounters considerable challenges in such networks owing to the high data and device heterogeneity. To address these challenges, we propose FedCLF, i.e., FL with Calibrated Loss and Feedback control, which introduces calibrated loss as a utility in the participant selection process and a feedback control mechanism to dynamically adjust the sampling frequency of the clients. The envisaged approach (a) enhances the overall model accuracy in case of highly heterogeneous data and (b) optimizes the resource utilization for resource constrained IoV networks, thereby leading to increased efficiency in the FL process. We evaluated FedCLF vis-à-vis baseline models, i.e., FedAvg, Newt, and Oort, using CIFAR-10 dataset with varying data heterogeneity. Our results depict that FedCLF significantly outperforms the baseline models by up to a 16% improvement in high data heterogeneity-related scenarios with improved efficiency via reduced sampling frequency.

Problem

• Addresses participant selection challenges in federated learning for IoV networks
• Improves model accuracy under high data heterogeneity conditions
• Optimizes resource utilization through dynamic client sampling adjustment

Innovation

• Uses calibrated loss for participant selection
• Employs feedback control to adjust sampling frequency
• Enhances model accuracy in heterogeneous data scenarios