Berlin 2018 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
DY: Fachverband Dynamik und Statistische Physik
DY 37: Networks: From Topology to Dynamics (joint session SOE/DY/BP) (joint session SOE/CPP/BP/DY)
DY 37.11: Talk
Wednesday, March 14, 2018, 12:00–12:15, MA 001
Non-inertial reference frames for inferring networks from dynamics — •Jose Casadiego1,2 and Marc Timme1,2 — 1Chair for Network Dynamics, Institute of Theoretical Physics and cfaed - Center for Advancing Electronics Dresden, Technical University of Dresden, Dresden, Germany — 2Network Dynamics, Max Planck Institute for Dynamics and Self-Organization (MPIDS), Goettingen, Germany
The dynamics of complex networks are determined to a great extent by the connectivity of their units. Given that measuring the connectivity by direct methods is often infeasible, researchers typically apply inverse approaches to infer links between units from the collective dynamics. Current state-of-the-art methods rely on either (i) quantifying functional links through statistical dependencies, or (ii) approximating the possibly nonlinear interactions between units via modeling of differential equations. Yet, functional links frequently do not match physical links, and finding an appropriate model may be computationally demanding and also require a prior knowledge about the interactions. Here we develop a model-independent theory to reconstruct the connectivity of networks from transients states to stable dynamics. Specifically, we demonstrate that representing these transients with respect to non-inertial reference frames provides simple linear mappings between network connectivity and dynamics. Furthermore, we show the robustness of our framework by reconstructing the full connectivity of different network dynamical systems exhibiting phase-locking, periodic orbits and collective synchronization.