Networks of intelligence. Networks of people.
We study both: AI systems that learn and operate without a central point of control, and the social systems — online and offline — that people build and inhabit. The two shape each other constantly, and that's the question our work sits inside.
IIT‑CNR, Pisa — Istituto di Informatica e Telematica
Research areas
all research →Decentralized learning
Training models across many devices without centralizing raw data — coordination-free protocols, robustness to node and data disruption, and the effect of network topology on decentralized federated learning.
Edge & pervasive AI
Bringing inference and learning onto resource-constrained and mobile devices, close to where data is generated.
Computational social science
Mathematical and LLM-based modeling of human behavior — social search and expert finding, personality prediction from text, and how real-world events reshape online social structure.
Network science
Structure and dynamics of complex networks — signed relationships, ego networks, and opinion dynamics — and what they reveal about how influence and information spread.
Causal learning
Discovering cause-effect structure from data under real-world constraints, including budgeted-intervention methods for pervasive and social systems.
Metaverse & virtual social networks
How social interaction and network structure change in immersive virtual environments, including AI-driven independent avatars that manage social interaction on a user's behalf.
Human-AI collaboration
Studying how humans and AI agents can work together effectively, including neurosymbolic approaches for human-in-the-loop settings.
Mobile & opportunistic networking
Communication and coordination among devices that connect intermittently, without relying on fixed infrastructure.