Mobility-Aware Multi-Task Decentralized Federated Learning for Vehicular Networks: Modeling, Analysis, and Optimization

2025-03-09

AI Summary

To address resource allocation challenges in vehicular networks—arising from high vehicle mobility, resource constraints, and concurrent multi-task workloads—this paper proposes a mobility-aware decentralized multi-task federated learning framework that jointly optimizes task scheduling, subcarrier allocation, and leader selection. We formulate the problem for the first time as a resource game with provable Nash equilibrium and innovatively recast it as a decentralized partially observable Markov decision process (DEC-POMDP). To solve it, we design a heterogeneous-agent proximal policy optimization (HAPPO) algorithm. We theoretically derive a convergence bound for single-task training, ensuring stability. Experimental results demonstrate that, compared to baseline methods, our approach improves model accuracy by 12.7%, reduces communication overhead by 28.4%, and increases task completion rate by 31.5%.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Planning, Routing, and Scheduling: Planning/Scheduling and Learning
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems

Abstract

Federated learning (FL) is a promising paradigm that can enable collaborative model training between vehicles while protecting data privacy, thereby significantly improving the performance of intelligent transportation systems (ITSs). In vehicular networks, due to mobility, resource constraints, and the concurrent execution of multiple training tasks, how to allocate limited resources effectively to achieve optimal model training of multiple tasks is an extremely challenging issue. In this paper, we propose a mobility-aware multi-task decentralized federated learning (MMFL) framework for vehicular networks. By this framework, we address task scheduling, subcarrier allocation, and leader selection, as a joint optimization problem, termed as TSLP. For the case with a single FL task, we derive the convergence bound of model training. For general cases, we first model TSLP as a resource allocation game, and prove the existence of a Nash equilibrium (NE). Then, based on this proof, we reformulate the game as a decentralized partially observable Markov decision process (DEC-POMDP), and develop an algorithm based on heterogeneous-agent proximal policy optimization (HAPPO) to solve DEC-POMDP. Finally, numerical results are used to demonstrate the effectiveness of the proposed algorithm.

Problem

• Optimize resource allocation in vehicular networks for multi-task federated learning.
• Address mobility and resource constraints in decentralized federated learning.
• Develop algorithms for efficient task scheduling and model training convergence.

Innovation

• Mobility-aware multi-task decentralized federated learning framework
• Resource allocation game with Nash equilibrium proof
• HAPPO algorithm for DEC-POMDP optimization