Detall
Dates:
Horari:
Organitza:
Lloc:
Afegeix-ho a l'agenda (iCal)
Conferenciant: Antoine Allard (Université Laval, Canada).
Resum: A central topic in network science is the study of the interplay between the global behavior of complex systems and the structure of the local interactions among their constituents. In this talk, I will briefly present two recent projects in which we develop methods to study this structure-behavior relationship in complex systems. The first project uses spectral theory to effectively reduce the dimension of the system of coupled ODEs describing the evolution of a dynamical process taking place on a network [1]. The second project exploits the tools of information theory to unveil a non-reciprocal relation between how much knowing about the network structure informs us about the evolution of a dynamical process (predictability), and how much knowing about the dynamical process (i.e. time series) informs us about the network it is evolving on [2].
[1]: https://arxiv.org/abs/2206.11230
[2]: https://arxiv.org/abs/2206.04000