Rescale-invariant federated reinforcement learning for resource allocation in v2X networks

Xu, Kaidi; Zhou, Shenglong; Li, Geoffrey Ye

doi:https://www.dx.doi.org/10.48550/arXiv.2405.01961

File(s)

2405.01961v1.pdf (395.33 KB)

Preprint

OA Location

https://ieeexplore.ieee.org/Xplore/home.jsp

Author(s)

Xu, Kaidi

Zhou, Shenglong

Li, Geoffrey Ye

Type

preprint

Abstract

Federated Reinforcement Learning (FRL) offers a promising solution to various practical challenges in resource allocation for vehicle-to-everything (V2X) networks. However, the data discrepancy among individual agents can significantly degrade the performance of FRL-based algorithms. To address this limitation, we exploit the node-wise invariance property of ReLU-activated neural networks, with the aim of reducing data discrepancy to improve learning performance. Based on this property, we introduce a backward rescale-invariant operation to develop a rescale-invariant FRL algorithm. Simulation results demonstrate that the proposed algorithm notably enhances both convergence speed and convergent performance.