Module

Voraussetzungen

Inhalte

• Unsupervised learning, strategy learning (Q-learning) • Markov decision processes (MDP) • Partially observable Markov decision processes (POMDP) • Bidirectional signal processing • Semantic modeling with feature-value relations • Petri-net transducers as semantic carriers • Modeling of higher cognitive processes (e.g. coping) • Applications in communications and dialog systems

Lernziele

After successfully completing the module, students can understand and develop techniques for semantics processing, methods for automatic planning and decision making under uncertainties, and the behavior control of cognitive technical systems.

Prüfungsleistung

• Processing of a seminar task in groups and presentation, approx. 10 minutes with subsequent technical discussion (25 %) (Solution of a programming task on the topic of the lecture and preparation of a presentation in self-study; presentation scheduled in the course of the lecture, usually on the last exercise date) • written examination, 60 minutes (75 %)

Bemerkungen

• Study programme Artificial Intelligence M.Sc.: Compulsory elective module in complex „Learning and Reasoning“ • Study programme Künstliche Intelligenz Technologie M.Sc.: Compulsory elective module in complex „Kognitions- und Neurowissenschaft“ • Study programme Informatik M.Sc.: Compulsory elective module in field of application „Maschinenbau/Elektrotechnik“ • Study programme Mathematics M.Sc.: Compulsory elective module in complex „Applications: Computer Science & Artificial Intelligence“