Artificial Intelligence

Description
Software systems which employs a combination of methods and techniques from Artificial Intelligence after the development of hybrid connectionist-symbolic models. Every natural intelligent system is hybrid because it performs mental operations on both the symbolic and subsymbolic levels. For the past few years there has been an increasing discussion of the importance of such an integration that led to the development of new simple and specific AI systems (such as systems for computer vision, speech synthesis, etc., or software that employs some of the models mentioned above). Now it’s time for integration to create broad AI systems to be employed in fields as:

• surveillance
• Ambient Assisted Living (AAL)
• Real-Time pattern-based decisions

“Istituto di Scienze e Tecnologie Cognitive”-CNR, Via San Martino della Battaglia 44, 00185 Rome

Aldo Franco Dragoni, Tel.: +39-0712204390, Email: a.f.dragoni@univpm.it, (Associate Professor)

Aldo Franco Dragoni was born in Ascoli Piceno (Italy), the 22nd of june 1961. Married since ‘92, he has two sons. He received a degree in Electronics Engineering from the University of Ancona, discussing a thesis in the field of Artificial Intelligence on “Plan Recognition from Visual Information”. Currently he is in charge as Associate Professor at the “Università Politecnica delle Marche”, where he teaches "Foundamentals of Computer Sciences", "Artificial Intelligence" and "Real Time Systems". His scientific interests derived from his PhD Thesis and regarded two questions. 1. The problem of belief revision in a multi-agent environment; if an agent receives new information that contradicts its previously held knowledge, how could it rearrange its corpus of beliefs in order to get rid of the contradiction? Aldo defined a model for belief revision in a multi-agent environment that integrates symbolic approaches to consistency maintenance with numerical methods to reason under uncertainty. The model is currently tested into a system to support detective inquiries. 2. The problem of plan recognition: if an agent observes another agent executing its own plan, how could it infer the other’s goals before the plan is completely executed? Aldo generalized the concept of plan recognition from observation to the idea of “mental states recognition” from communication, by defining three abductive methods to infer a speaker’s mental state from his utterances (“speech acts”). These methods enrich the huge literature on user modeling in intelligent user interfaces, and constitute another knowledge acquisition medium. Aldo devoted much interest also in the study of distributed information systems, under the realistic hypothesis that they can be affected by limited degrees of correctness and completeness. He tried to evaluate, on statistical simulation basis, the global performances of a group whose members adopt the same local strategies for belief revision, the same policies for communication and the same criteria for partner selection. He served as Program Committee and Reviewer for several International Conferences and Journals on Artificial Intelligence.