• Decentralized Federated Learning
• Echo State Networks
• Federated Learning
• Pervasive Computing

Federated Echo State Networks proved their efficiency in learning low-resource collaborative settings where data is regulated privacy. In this thesis, we broaden the applicability of this machine learning approach to a decentralized setting, where multiple agents collaborate to learn a global readout with a one-shot, exact consensus mechanism. Experiments prove the efficacy and the efficiency of the proposed learning methodology against a state-of-the-art competitor on multiple benchmarks, characterized by different levels of statistical heterogeneity.