Bayesian Modeling and Information Theory for neuroscience and cognitive science


Eugenio Piasini

Amount of frontal teaching: 

~16 hours

Ideas at the interface of information theory and Bayesian statistics have long been a source of inspiration and of powerful quantitative methods in neuroscience and cognitive science. This course will focus on some key foundational concepts, and how they are used to formulate and test theories of cognition, perception, and neural processing.

Rough syllabus: (i) Introduction to Bayesian inference; (ii) Perception as Bayesian inference; (iii) Bayesian inference under sensory noise; (iv) Cue combination and evidence accumulation; (v) Stimulus detection, discrimination and classification; (vi) Introduction to information theory; (vii) The efficient coding principle.