Especially we focus on the establishment of an online training environment based on RIC platform. We used various open sources for serving ML model as a inference service and building an ML training pipeline for pipeline automation for online training. We trained a sample reinforcement learning (RL) model which controls function parameters in Distributed Unit (DU). After training it with data from a specific cell then its deployed to a different environment. To optimize the model performance, we execute the training pipeline for retraining the model using online workflow, then get better result

https://arxiv.org/abs/2207.04949