Uso de cookies

En las páginas web de la Universidad Carlos III de Madrid utilizamos cookies propias y de terceros para mejorar nuestros servicios mediante el análisis de sus hábitos de navegación. Al continuar con la navegación, entendemos que se acepta nuestra política de cookies. "Normas de uso"

Carlo Regazzoni


Carlo Regazzoni - Universita Degli Studi di Genova, Italy

Carlo S. Regazzoni (Ph.D. 1992) He is full professor at DITEN Dept., University of Genova, Italy, where he is head of the Information and Signal Processing for Cognitive Telecommunications research group. Since 2009 he is coordinator of international PhD courses (EMJD ICE, ICE) on Interactive and Cognitive Environments. His research interests include ( cognitive dynamic systems, data fusion, adaptive and self-aware video and signal processing, machine learning, software and cognitive radios. He is author of books (4), peer-reviewed papers on international journals (90) and international conferences (340). He served as General Chair, Technical program chair and in other committees of several international conferences; associate/guest editor of international journals (IEEE Trans on: Image Processing, Circuits and Systems for Video Technology, Proc. of the IEEE, IEEE Signal Processing Magazine et al.). He served in IEEE Signal Processing Society in multiple roles including as Vice President Conferences in 2015-2017.



The scientific goal of the project is to study and develop behavioral self-aware representation and inference techniques to allow embodied Cognitive Dynamic System associated with physical autonomous agents to learn from experiences observed and proactive situation patterns. Such representations should be capable to incrementally evolve based on experiences which the agent is exposed to. Moreover they should allow to relate agent sensory observations oriented to external context with sensory observations monitoring the agent state itself, so allowing self awareness. Uncertainties of sensory observations will have to be modeled as well causal and temporal relations of observations with other parts of representations, e.g. state estimations at continuous and semantic (discrete) level. Therefore, dynamic probabilistic models suited to non stationary domains will be studied and, in particular, non parametric techniques. Machine learning techniques like imitation and transfer learning will be considered to develop innovative representations in applications like autonomous driving, smart buildings surveillance and cognitive radio

Imagen Cátedras de Excelencia 2014