Reducing Non-IID Effects in Federated Autonomous Driving with Contrastive Divergence Loss

Abstract-- Federated learning has been widely applied in autonomous driving since it enables training a learning model among vehicles without sharing users' data. In this paper, we propose a new contrastive divergence loss to address the non-IID problem in autonomous driving by reducing the impact of divergence factors from transmitted models during the local learning process of each silo. We also analyze the effects of contrastive divergence in various autonomous driving scenarios, under multiple network infrastructures, and with different centralized/distributed learning schemes. Autonomous driving is an emerging field that enables On the other hand, DFL does not require a server and uses vehicles to operate without a human driver by using a a fully distributed network. In autonomous driving, several combination of vision, learning, and control algorithms to works have explored both DFL and SFL to address different observe and respond to changes in the environment [1].