Learning from data: Applications of Machine Learning in optical network design and modeling

This article overviews the uses and applications of classical Machine Learning techniques in a variety of network problems. We first overview the basics of statistical learning, including the main algorithms and methodologies involved in the process of designing good Machine Learning models. The second part addresses a number of network use cases where ML can be used to complement and extend existing network models and algorithms, including the classical Routing and Wavelength Assignment (RWA) problem and fiber access delay modelling.