Abstract
This master's project aims to design and evaluate federated learning models that fit telecommunications operator scenarios, investigating which distributed learning architectures, such as vertical, horizontal, hybrid, or hierarchical federated learning, best represent the requirements and needs of applications in such scenarios.