| The free energy landscape of a spin glass has a very rugged character due to the random spin couplings. Ideally we would like to be able to find low-lying minima, or at the very least, understand how the a particular set of dynamical rules influences the character of the minima that they are attracted to. In this talk, I will discuss various approaches to finding free energy minima in two mean-field formulations of the Sherrington-Kirkpatrick spin glass, and speculate as to the potential general utility of the algorithms in the wider context of optimisation problems. |
|