NeFF − Terminankündigung / Einladung
LOEWE-NeFF and the Frankfurt Institute for Advanced Studies (FIAS)
jointly invite to a
Workshop on Neural Information Dynamics, Causality and Computation close to Criticality
December 12-13th, 2014
The workshop is preceded by a
Student Course on Neural Information Dynamics with TRENTOOL, the Java Information Dynamics Toolkit and MuTE
December 10-11th, 2014
Venue: Workshop and student course will be held at the Frankfurt Institute for Advanced Studies (FIAS, www.fias.uni-frankfurt.de), Ruth-Moufang-Straße 1, 60438 Frankfurt, Germany
The workshop addresses the analysis of neural computation in large neural systems and covers three tightly related topics in the fields of modern analysis of neural data:
The analysis of causal interactions yields important insights into the biophysical substrate of neural dynamics that enable emergent computation. It is one of the goals of the workshop to discuss the link between causal analysis and the analyses of information processing proper, in order to clarify the dividing line and the mutual benefit of these two types of analyses.
Neural information dynamics
Information theoretic quantities separate and measure key elements of distributed computation in neural systems, such as the storage, transfer, and modification of information. These concepts can help to better understand the computational algorithms implemented by the dynamics of a neural system, providing the link between these algorithms and their biophysical implementation.
Large scale organisation and criticality
Neural systems orchestrate the activity of an enormous number of interacting neurons to achieve their computational capabilities. Recent advances in recording technology make it possible to reveal the activity patterns of an ever increasing number of neurons. This offers the unique opportunity to identify the large-scale organizing principles of neural activity, such as operation near a critical point, that support computation in the system.
The student course addresses those intending to apply information theoretic maesures for neuroscience hands on, and those that would like to contribute code to one of the open source toolboxes on the topic. Some minimal experience with MATLAB or GNU-Octave is necessary for
participation. Some prior knowledge in information theory is a plus. Applications for the student course should include a brief statement of motivation.