The goal of Artificial Intelligence is to give machines a certain "intelligence" = ability to "reason" = to deduce and to induce. This involves the manipulation of knowledge: the representation and application of information relating to the problem to be solved. In addition, AI can be studied on 2 main axes: classic AI where the machine is limited to deducing knowledge from those acquired (reasoning, modern expert systems, etc.); and more recent AI where the machine seeks to induce knowledge from examples (cf. Artificial learning). This course gives priority to classical AI and presents the basic techniques and tools used in reasoning and problem solving in different areas of AI. The study and implementation of these techniques and tools call upon advanced concepts such as graphs and objects. Some basic notions in logic will be presented in order to facilitate the use of the Prolog language allowing logical reasoning. Examples of industrial use of intelligent systems - experts will be presented. Also, this course introduces the "constraint programming" paradigm which uses logic together with (numerical) constraints to provide a powerful tool for reasoning AND optimization.
1 - Outils et Techniques de Représentation de Connaissances 2 - Techniques de Manipulation de Connaissances 3 - Outils et Langages 4 - Systèmes à base de règles 5- Logique et Contraintes : modélisation de problèmes complexes 6- Interrogation de bases de données relationnelles avec Prolog 7 - Introduction : Tableau de bord, systèmes à base d'agents, Apprentissage, etc. En parallèle, les élèves choisiront un sujet à préparer en groupe avec restitution et rapport final.