Course details


Bayesian Data Analysis: Advanced Topics

SS 2020 Prof. Dr. Michael Franke, Dr. phil. Timo Röttger
B.Sc modules:
CS-BWP-AI - Artificial Intelligence
CS-BWP-CNP - Cognitive (Neuro-)Psychology
CS-BWP-MCS - Methods of Cognitive Science
KOGW-WPM-KI - Artificial Intelligence
KOGW-WPM-KNP - Cognitive (Neuro-)Psychology
M.Sc modules:
CC-MWP-AI - Artificial Intelligence
CS-MWP-AI - Artificial Intelligence
CS-MWP-CNP - Cognitive (Neuro-)Psychology
KOGW-MWPM-KI - Major subject Artificial Intelligence
KOGW-MWPM-KNP - Major subject Cognitive (Neuro-)Psychology
Tue: 16-18
Thu: 14-16

Prerequisites: Statistics & Data Analysis (or equivalent) Machine Learning (or equivalent) Firm command of R or Python (ideally both) The course introduces probabilistic programming tools for Bayesian inference (Stan, WebPPL, possible Pyro,... ) together with standard algorithms for approximating a posterior distribution (variational inference, MCMC, ...). It visits Bayesian analyses of selected cognitive models (signal detection, drift diffusion, ...), linguistic models (Rational Speech Act, ...) and machine learning models (clustering, topic, ...). This is an advanced course. Students are expected to have prior knowledge of the basics of Bayesian Data Analysis.