The research line of the group focuses on the design and development of learning strategies inspired by human cognitive mechanisms for Artificial Intelligence. The aim is to translate into artificial systems the principles of adaptation, plasticity, and consolidation typical of the human brain, translating them into Continual and Lifelong Learning approaches capable of acquiring new knowledge without catastrophic forgetting. In parallel, the group explores Federated Learning methodologies to enable distributed and collaborative learning accross multiple entities without direct data sharing, complemented by unlearning techniques for the targeted removal of obsolete or undesired knowledge. Activities ecompass both the definition of novel bio-inspired paradigms and the implementation of efficient, scalable, and trustworthy models, designed to operate in dynamic and heterogeneous contexts. The overarching goal is to advance toward AI systems that are more flexible, resilient, and ethically responsible.