• federated echo state networks
• echo state networks
• federated learning

Nowadays Internet of Things (IoT) provides ubiquitous sensing and computing capabilities. Artificial Intelligence provides insights on the data generated by IoT devices. At the same time the training data with sensitive user informations raises privacy concerns. So, the necessity for developing privacy-enhanced intelligent IoT networks and applications. In the context of the Human State Monitoring task of the TEACHING project, We experiment learning models that let the cyber-physical systems of systems be humanistic intelligent, and so able to adapts itself by means of human feedback. We train an Echo State Network ables to recognize the stress of a human driver. Then, we analyze such model in a Federated Learning scenario and we use and compare different federated learning algorithms.