Gabriele Tolomei is an Associate Professor of Computer Science
at Sapienza University of Rome, where he leads the
HERCOLE Lab
(Human-Explainable, Robust, and COllaborative LEarning).
He is also co-founder and Chief Scientific Officer at tellmewAI, an innovative start-up developing tools for the automatic compliance verification of AI systems against regulatory frameworks, such as the European AI Act.
His research is centred on three core pillars of modern AI: explainability,
security, and scalability.
In explainability, his work focuses on post-hoc,
model-agnostic methods—particularly counterfactual explanations—for complex, "black-box"
machine learning systems.
The goal is to make AI decisions transparent, actionable, and human-understandable.
In security, he studies adversarial machine learning,
analysing vulnerabilities of models under evasion and poisoning attacks,
and designing methods to improve the robustness and reliability of both
centralized and federated learning systems.
In scalability, his research addresses large-scale and
distributed learning, including federated learning, with a focus on efficiency, privacy, and deployment
in real-world settings.
Before joining Sapienza, he held research positions at Yahoo Labs (London, UK) and
ISTI-CNR (Pisa, Italy), working on advertising quality, query understanding, and large-scale data mining.
