Parts | Days | Selection | Search | Downloads | Help
BP: Fachverband Biologische Physik
BP 26: Posters II
BP 26.50: Poster
Thursday, February 28, 2008, 17:00–19:30, Poster A
Which Network Connectivities generate a given Dynamics? I: Optimal Network Design — •Raoul-Martin Memmesheimer1,2, Lishma Anand1,2, Frank van Bussel1,2, and Marc Timme1,2 — 1Network Dynamics Group, MPI f. Dynamics & Self-Organization — 2Bernstein Center for Computational Neuroscience, Göttingen
We present alternative perspectives towards understanding relations between structure and dynamics in neural networks.
First, can we design a network, e.g. by modifying the features of units or interactions, such that it exhibits a desired dynamics? Here we positively answer this question analytically for a class of networks of spiking neural oscillators [1, 2], by finding the set of all networks that exhibit a given arbitrary periodic spike pattern as an invariant dynamics. We illustrate the applicability of the method by designing networks that exhibit a predefined dynamics and simultaneously minimize the networks' wiring costs, i.e. are structurally optimal.
In a second contribution "Network Reconstruction" we present a method to infer the connectivity of a given network from its response dynamics to external driving signals [3,4].
[1] R.-M. Memmesheimer and M. Timme,
Phys. Rev. Lett. 97:188101 (2006).
[2] R.-M. Memmesheimer and M. Timme, Physica D 224:182 (2006).
[3] M. Timme, Europhys. Lett. 76:367 (2006).
[4] M. Timme, Phys. Rev. Lett. 98:224101 (2007).