Luciano Serafini

Head of Unit

    Short bio

    Luciano Serafini is a researcher in the area of Artificial Intelligence. He is currently the head of the Data and Knowledge Management Research Unit at Fondazione Bruno Kessler. He graduated in information science with a thesis on logic for knowledge representation. Since 1990 he joined Fondazione Bruno Kessler (before ITC-IRST) as a researcher in the area of logic for knowledge representation and reasoning.  During more than 30 years of research career he contributes with few important ideas in the field of Artificial Intelligence.

    In the 90's he contributed to the definition of a formalism called Multi-Context (MC) System for the representation of modular interconnected context-dependent knowledge. MC Systems has a great influence on the semantic web area and in the area of information integration. Between 2000-2010 he worked on semantic matching for the integration of heterogeneous schemas with other researchers in Trento who suggested encoding the problem of matching heterogeneous hierarchical classifications in terms of propositional satisfiability. After 2010, Luciano Serafini started his research about integrating machine learning and logical reasoning, and in 2016 he contributed to the definition of  Logic Tensor Network, one of the first neuro-symbolic architectures. In the last few years, he also has been interested in integrating learning, acting, and planning. His current interests include Embodied Artificial Intelligence, Neuro-Symbolic Integration, Statistical Relational Learning, and integrating Planning, Acting, Perception, and Reasoning. From 2020 he is an EurAI fellow and teaches regular courses on Knowledge representation and learning at the University of Padova.

    Activities

    Research topics


    • First Order Weighted Model Counting: I'm interested in developing computationally efficient techniquest for weighted model counting in fragments of first order logic extended with additional axioms.

    • Integrating Planning, Acting, and Learning: Development of cognitive architectures and algorithms that allows the tight integration of planning, acting, perceiving, learning and reasoning. These type of architectures should suppor the development of Embodied AI agents that are capable to operate and learn in open enviroments.

    • Logics for distributed knowledge representation and reasoning. Developement of formal models for integrating different knowledge representation systems.

    • Neuro simboli intrgration: Study how knowledge representation frameworks based on logic, can be integrated with subsimbolic models such as probabilistic models and neural networks.

    Latest news