---

The inverse Ising problem seeks to reconstruct the parameters of an Ising Hamiltonian on the basis of spin configurations sampled from the Boltzmann measure. The last decade has witnessed many applications of the inverse Ising problem driven by the advent of large-scale data in different scientific disciplines, especially biology. In this talk I will give an introduction to the inverse Ising problem, discussing its applications as well as different approaches from statistical physics. A particular focus is on the link between approaches to the inverse Ising problem based on the optimisation of some `objective function' and the statistical physics of disordered systems. This allows to characterise the performance of an arbitrary objective function and to calculate the objective function which optimally reconstructs the model parameters.

---

Johannes Berg, University of Cologne, Institute for Theoretical Physics
TP seminar room 0.03
Contact: not specified