Formulating QoT Estimation with Machine Learning

We investigate two machine learning approaches to formulate QoT estimation where we use monitoring information to train i) the inputs of an analytical physical layer model or ii) a machine learning model. We study their suitability for various uncertainty scenarios.