Decentralized Rank Scheduling for Energy-Constrained Multi-Task Federated Fine-Tuning in Edge-Assisted IoV Networks

2025-08-13

AI Summary

To address low multi-task federated fine-tuning efficiency and high latency in edge-assisted Internet of Vehicles (IoV) caused by high vehicle mobility, resource heterogeneity, intermittent connectivity, and energy constraints, this paper proposes a hierarchical federated fine-tuning framework integrating roadside units (RSUs) and vehicles for collaborative multi-task LoRA-based low-rank adaptation. We innovatively design a decentralized energy-aware rank scheduling mechanism, formulated as a multi-armed bandit problem, and propose the UCB-DUAL algorithm to jointly optimize exploration-exploitation trade-offs and dynamic energy constraints, achieving sublinear regret guarantees. Extensive evaluations on a large-scale, real-trajectory-driven IoV simulation platform demonstrate that our approach improves average accuracy by over 2.5% and reduces end-to-end latency by 24% compared to state-of-the-art baselines, significantly enhancing the accuracy–energy-efficiency trade-off.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Distributed Machine Learning & Federated Learning
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Federated fine-tuning has emerged as a promising approach for adapting foundation models (FMs) to diverse downstream tasks in edge environments. In Internet of Vehicles (IoV) systems, enabling efficient and low-latency multi-task adaptation is particularly challenging due to client mobility, heterogeneous resources, and intermittent connectivity. This paper proposes a hierarchical federated fine-tuning framework that coordinates roadside units (RSUs) and vehicles to support resource-aware and mobility-resilient learning across dynamic IoV scenarios. Leveraging Low-Rank Adaptation (LoRA), we introduce a decentralized, energy-aware rank adaptation mechanism formulated as a constrained multi-armed bandit problem. A novel UCB-DUAL algorithm is developed to enable adaptive exploration under per-task energy budgets, achieving provable sublinear regret. To evaluate our method, we construct a large-scale IoV simulator based on real-world trajectories, capturing dynamic participation, RSU handoffs, and communication variability. Extensive experiments show that our approach achieves the best accuracy-efficiency trade-off among all baselines, reducing latency by over 24% and improving average accuracy by more than 2.5%.

Problem

• Efficient multi-task federated fine-tuning in dynamic IoV networks
• Energy-aware rank adaptation under constrained resources
• Mobility-resilient learning with decentralized scheduling

Innovation

• Hierarchical federated fine-tuning framework for IoV
• Decentralized energy-aware rank adaptation mechanism
• UCB-DUAL algorithm for adaptive energy-efficient exploration