Federated Learning for INdustrial Twins
Federated, edge-based AI integrating Digital Twins for privacy-preserving intelligence in critical sectors.
People & community
M.O.D.A.L. Research Group
Artificial Intelligence, data analysis and mathematical modelling at the University of Naples Federico II.
Institutional identityM.O.D.A.L. · DMA · University of Naples Federico IIMathematical mOdelling and Data AnaLysis
M.O.D.A.L. @ UNINAResearch
AI grounded in science,
built for real systems
Machine & Deep Learning
Learning architectures and data-driven methods for complex, heterogeneous and large-scale data.
Federated Learning
Distributed intelligence, non-IID data, systems heterogeneity, privacy and sustainable edge AI.
Scientific ML
Physics-informed neural networks and hybrid models that integrate data with scientific knowledge.
Digital Twins & HPC
Edge–Cloud–HPC ecosystems for simulation, monitoring, prediction and decision support.
New interactive research resource · 2026
Atlante PRIN 2026
Explore Italy’s PRIN 2026 research landscape through an interactive geographic atlas developed by M.O.D.A.L. Browse projects and research areas across the standard, Hybrid and Synergy programmes.
Latest · 2026
News & events
Recent research results, scientific events and laboratory activity.
Prof. Zenglin Xu visits M.O.D.A.L.
Invited seminar at Federico II: Trustworthy Federated Learning: Toolboxes, Benchmarks, and Models.
ECML PKDD 2026: FLIP4Tab & AGENSYS
Main-track research and workshop activity as the international ML community met at Federico II.
University of Tokyo visit to M.O.D.A.L.
Exchange on AI, urban data and future research directions with Prof. Hangli G.
FuGuard: federated unlearning
Generative surrogates and optimal transport for client-level machine unlearning in federated systems.
Featured projects
Research in motion
A selection of current initiatives across European, national and regional programmes.
PN RIC · 2026ECHO-TWIN
Edge-Cloud-HPC Optimised Twins
An integrated ecosystem for Digital Twin-enabled applications across health, climate, mobility and AI.
Towards an Understanding of Artificial Intelligence
Training 13 doctoral candidates in sustainable, transparent, robust and human-centred AI.
People
A multidisciplinary research group
Computer science, numerical analysis, scientific computing and advanced AI in one collaborative research environment.
Francesco PiccialliFounder, Head & Scientific Coordinator · Computer Science
Salvatore CuomoCo-Founder · Scientific Machine Learning · Numerical Analysis
Marco LapegnaHigh-Performance Computing · Computer Science
Valeria MeleComputer Science · Artificial Intelligence
Faculty, researchers and Ph.D. students →
AIMachine & Deep Learning
FLPrivacy-preserving intelligence
HPCScalable scientific computing
DTDigital Twin systems
Selected publications
Recent research output
2026
FuGuard: Client-Level Federated Unlearning via Generative Surrogates and Optimal Transport
↗
2026
Only one fusion matters: Spatiotemporal Weighted Integration for one-shot Federated Traffic Prediction
↗
2026
Deep learning framework for determining earthquake focal mechanisms with application to the Campi Flegrei caldera
↗
Outreach & Media
M.O.D.A.L. in the public conversation
Expert commentary, essays and press features connecting research with the broader debate on AI and data-driven society.
Featured essay · Civiltà dei Dati
Fabbriche e manifatture
Francesco Piccialli on intelligent Digital Twins, AI and the transformation of industrial systems.
Read at Fondazione Leonardo ↗Quando l’IA impara a dimenticare
Machine unlearning, privacy, security and governance across the AI lifecycle.
Read article ↗Francesco Piccialli tra scienza e futuro
A long-form interview on M.O.D.A.L., research and the role of AI in society.
Read interview ↗Research, innovation and collaboration
M.O.D.A.L. works with universities, companies and public stakeholders on advanced AI, mathematical modelling and data-driven systems.