Strong replica symmetry for high-dimensional disordered log-concave Gibbs measures
This provides theoretical insights for machine learning and high-dimensional inference contexts where log-concave measures appear, such as in convex empirical risk minimization and M-estimation.
The paper tackles the problem of understanding the asymptotic behavior of high-dimensional disordered log-concave Gibbs measures by proving concentration of multioverlap order parameters, which leads to a simple representation of asymptotic Gibbs measures and strong decoupling of variables.
We consider a generic class of log-concave, possibly random, (Gibbs) measures. We prove the concentration of an infinite family of order parameters called multioverlaps. Because they completely parametrise the quenched Gibbs measure of the system, this implies a simple representation of the asymptotic Gibbs measures, as well as the decoupling of the variables in a strong sense. These results may prove themselves useful in several contexts. In particular in machine learning and high-dimensional inference, log-concave measures appear in convex empirical risk minimisation, maximum a-posteriori inference or M-estimation. We believe that they may be applicable in establishing some type of "replica symmetric formulas" for the free energy, inference or generalisation error in such settings.