Max Planck Institute for Dynamics and Self-Organization -- Department for Nonlinear Dynamics and Network Dynamics Group
Personal tools
Log in

BCCN/BFNT AG-Seminar

Tuesday, 02.07.2013 17 c.t.

Information theoretic principles in learning and decision making

by Prof. Tatyana Sharpee
from Computational Neurobiology Laboratory, The Salk Institute, La Jolla, CA, USA

Contact person: Dominika Lyzwa

Location

Ludwig Prandtl lecture hall

Abstract

In this talk I will discuss information theoretic principles for adaptation, learning and decision making. First I will describe how adapting properties of two types of retinal ganglion cells are coordinated to maximize information transmission. The emergence of these two types of cells can be understood as a symmetry breaking associated with a second-order phase transition. In the second part of my talk I will describe how search strategies by C. elegans can be mapped onto information maximization problem about their behavioral goals. The solution provides quantitative match between worm behavior the population level. Furthermore, it is possible to account for individual differences in the worm behavior by taking into account variations in their prior sensory experiences.

back to overview