Safe and Accelerated Deep Reinforcement Learning-based O-RAN Slicing: A Hybrid Transfer Learning Approach
This addresses practical deployment challenges for DRL in O-RAN slicing, offering incremental improvements for network operators.
The paper tackles slow convergence and unstable performance in deep reinforcement learning (DRL) for O-RAN slicing by proposing a hybrid transfer learning approach, achieving improvements such as a 64.6% decrease in reward variance and up to 20.7% increase in converged scenarios compared to baselines.
The open radio access network (O-RAN) architecture supports intelligent network control algorithms as one of its core capabilities. Data-driven applications incorporate such algorithms to optimize radio access network (RAN) functions via RAN intelligent controllers (RICs). Deep reinforcement learning (DRL) algorithms are among the main approaches adopted in the O-RAN literature to solve dynamic radio resource management problems. However, despite the benefits introduced by the O-RAN RICs, the practical adoption of DRL algorithms in real network deployments falls behind. This is primarily due to the slow convergence and unstable performance exhibited by DRL agents upon deployment and when encountering previously unseen network conditions. In this paper, we address these challenges by proposing transfer learning (TL) as a core component of the training and deployment workflows for the DRL-based closed-loop control of O-RAN functionalities. To this end, we propose and design a hybrid TL-aided approach that leverages the advantages of both policy reuse and distillation TL methods to provide safe and accelerated convergence in DRL-based O-RAN slicing. We conduct a thorough experiment that accommodates multiple services, including real VR gaming traffic to reflect practical scenarios of O-RAN slicing. We also propose and implement policy reuse and distillation-aided DRL and non-TL-aided DRL as three separate baselines. The proposed hybrid approach shows at least: 7.7% and 20.7% improvements in the average initial reward value and the percentage of converged scenarios, and a 64.6% decrease in reward variance while maintaining fast convergence and enhancing the generalizability compared with the baselines.