Regensburg 2016 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 3: Networks: From Topology to Dynamics I (joint session SOE / DY / BP)
SOE 3.1: Talk
Monday, March 7, 2016, 10:00–10:15, H36
Revealing physical interaction networks from nonlinear dynamics — Jose Casadiego1, Dimitra Maoutsa1, Hauke Hähne1, Mor Nitzan2, and •Marc Timme1 — 1Network Dynamics, MPI for Dynamics and Self-Organization, Göttingen, Germany — 2Racah Institute of Physics, The Hebrew University of Jerusalem, Israel
Structural connectivity of networks reflects the direct physical interactions between pairs of dynamical units, as opposed to effective, functional or other statistical measures of connectivity. How to uncover physical interaction structure from measured time series of networked systems remains an open question. Here we present a dynamical systems’ view on collective network dynamics, thereby proposing an approach to reveal physical interaction networks from the nonlinear dynamics they generate. Introducing the notion of explicit dependency matrices, we present two examples: one, where the time series consists of the full network states as a function of time, the other, where the time series exhibits only partial information about the full states. We apply the latter to neural circuit dynamics where the observables are spike timing data, i.e. only a discrete, state-dependent output of the neurons. These results may help revealing network structure for systems where direct access to dynamics is simpler than to connectivity.
For an introductory review of the state of the art, see
J. Phys. A: Math. Theor. 47 343001 (2014)
http://dx.doi.org/10.1088/1751-8113/47/34/343001