Machine Learning vs. Classical Methods: Evaluating Congestion Control Performance in the Transport Layer

Abstract

Over time, the Internet has integrated new communication technologies for stationary and mobile networks, resulting in improved performance and higher data rates. The Internet research community has worked diligently to enhance transport layer protocols and leverage machine learning to optimize network operations. This paper focuses on the innovation and contribution of learning-based congestion control approaches, particularly reinforcement learning, and discusses the current challenges and trends in this field.