Biography

PhD candidate at the M.O.D.A.L. group, her work focuses on developing adaptive algorithms for decentralized learning environments, particularly in Federated Learning. Her work addresses key challenges, such as handling data heterogeneity and privacy. With a growing reputation as an emerging researcher in AI and FL, she has contributed to several impactful publications, including in high-impact journals and top-tier conferences. A highlight of her recent work is the paper titled KAFE: Kernel Aggregation for FEderated, which has been accepted for presentation in the main track of the prestigious ECML-PKDD 2024. In addition, Diletta has demonstrated strong leadership in the academic community by organizing the Special Session on Federated Learning on Big Data at IEEE BigData 2024, which will be held in Washington DC, USA, in December 2024. She wishes to continue her contributions to advancing the frontiers of Federated Learning and its applications across various domains, aiming to address emerging challenges and push the boundaries of what decentralized learning systems can achieve.