Replica Symmetry Breaking Condition Exposed by Random Matrix Calculation of Landscape Complexity

TL;DR

This paper uses random matrix methods to analyze the landscape complexity of high-dimensional Gaussian energy landscapes, revealing conditions for replica symmetry breaking and the behavior of stationary points and minima.

Contribution

It introduces a random matrix approach to connect replica symmetry breaking with the exponential growth of stationary points and minima in Gaussian landscapes.

Findings

Replica symmetry breaking coincides with exponential number of stationary points and minima.

Complexity vanishes cubically near critical confinement, quadratically for cumulative complexity.

Near critical confinement, saddle points with positive complexity are close to minima with few negative Hessian eigenvalues.

Abstract

We start with a rather detailed, general discussion of recent results of the replica approach to statistical mechanics of a single classical particle placed in a random $N (\gg 1)$-dimensional Gaussian landscape and confined by a spherically symmetric potential suitably growing at infinity. Then we employ random matrix methods to calculate the density of stationary points, as well as minima, of the associated energy surface. This is used to show that for a generic smooth, concave confining potentials the condition of the zero-temperature replica symmetry breaking coincides with one signalling that {\it both} mean total number of stationary points in the energy landscape, {\it and} the mean number of minima are exponential in $N$. For such systems the (annealed) complexity of minima vanishes cubically when approaching the critical confinement, whereas the cumulative annealed complexity vanishes quadratically. Different behaviour reported in our earlier short communication [Y.V. Fyodorov et al., {\it JETP Lett.} {\bf 85}, 261, (2007)] was due to non-analyticity of the hard-wall confinement potential. Finally, for the simplest case of parabolic confinement we investigate how the complexity depends on the index of stationary points. In particular, we show that in the vicinity of critical confinement the saddle-points with a positive annealed complexity must be close to minima, as they must have a vanishing {\it fraction} of negative eigenvalues in the Hessian.