Inferring network structure using dynamical mean-field theory

Feb 24, 2011, 9:00 AM
FB42 (AlbaNova Main Building)


AlbaNova Main Building


Yasser Roudi (Norwegian University of Science and Technology)


I will describe how interactions in a non-equilibrium Ising model can be inferred from observing state samples. We will start by a short review of how this could be done for equilibrium systems and then study how Dynamical Mean-Field (naive mean field and TAP) theory can be developed for a nonequilibrium Sherrington-Kirkpatrick model and exploited for inferring the network connectivity. We will also quantify the error in inferring the connectivity in the high temperature regime.

Presentation materials

There are no materials yet.