University of Chicago, July 25, 2002
Complex nonlinear neural dynamics.
Peter Andras (Newcastle, UK)
Robert Kozma (Memphis, TN)
Amir Assadi (Wisconsin, WS)
Mini-symposium, full day
Studying nonlinear, dynamical aspects of information processing and memory in biological and computational neural systems has attracted a lot of attention in recent years. This session aims at giving a forum to researchers interested in this field introducing their results and discussing new developments in this area. The major questions addressed at this session will be: What evidence do we have that indicates that neurons, neural circuits or brain components have functionally meaningful complex nonlinear dynamic or chaotic activity? How should we interpret these experimental results? What is the role of noise in the operation of biological systems? How does noise contribute to the stability of complex nonlinear systems in biology and in computational models? What are the most promising approaches to the analysis of complex nonlinear dynamics in neural network models? What spatio-temporal features indicate complex behavior? What new insights are provided by the analysis of lattices of coupled nonlinear maps? Is symbolic dynamics the right conceptual framework to describe the dynamical macro-behavior of neural systems? How does the chaotic dynamics analysed in the context of some neural network models correspond to the complex spatio-temporal dynamical effects observed in biological brains ? What are the established links between the two kinds of complex neural dynamics and what are the gaps that are waiting to be bridged? Is it essential to have complex nonlinear dynamics in the neural system to perform computation or would it be more advantageous to do it without any chaos? Is it possible that chaos is just a side effect from the point of view of neural computation?
More information is available from www.staff.ncl.ac.uk/peter.andras/cnnd-ws2002.htm.
The workshop will consist of a few invited talks, a selection of short talks and discussion panels. We invite participants of the CNS 2002 to submit proposals for short talks at the workshop. The proposals should be in the form of a one page abstract, summarizing the content of the talk and indicating which of the above listed questions are going to be addressed by the talk. Please indicate also if the proposed talk is different from your paper(s) submitted to the CNS 2002. The selected proposals will be allocated a 10 minutes presentation time. All other proposals will be considered for presentation as posters. Abstracts should be submitted by e-mail to the address: firstname.lastname@example.org. Deadline for abstracts: July 3, 2002. E-mail notification about acceptance for oral presentation will be sent out by July 12, 2002. The organizers intend to publish a volume of the papers related to the talks given at the workshop.