Building upon the lecture "Cognitive Computing: Reasoning and Decision Making under Uncertainty", in this seminar we will explore how Bayesian Models can be and have been used in real world scenarios. We try to discuss how suitable different approaches are for different scenarios and look into what is required to actually implement these models successfully. This also includes discussion fundamental requirements of computationally dealing with different kinds of distributions and the approximations required in the binary system of out computers.
The models under discussion will be jointly decided in the first session, potential candidates include (Dynamic) Bayesian Networks, (Hidden) Markov Models, Probabilistic Neuronal Networks, (Hierarchical) Gaussian Processes and Bayesian Partition Models.
This seminar aims to provide both theoretical discussions about different methods as well as some hands on experience with actually implementing such methods.
Teilnahmevoraussetzungen, notwendige Vorkenntnisse
Lecture "Reasoning and Decision Making unter Uncertainty"