Rarity of extremal edges in random surfaces and other theoretical applications of cluster algorithms

TL;DR

This paper demonstrates that extremal edges are rare in random surfaces under Lipschitz constraints, extending previous results to general graphs using cluster algorithms, and applies these findings to spin models and correlation inequalities.

Contribution

It introduces a new proof technique using cluster algorithms applicable to general graphs, broadening the understanding of extremal edges in random surfaces.

Findings

Extremal edges are rare in Lipschitz-constrained random surfaces.

Cluster algorithms can be effectively used for theoretical analysis of random surfaces.

Monotonicity of pair correlation densities in the spin O(n) model is established.

Abstract

Motivated by questions on the delocalization of random surfaces, we prove that random surfaces satisfying a Lipschitz constraint rarely develop extremal gradients. Previous proofs of this fact relied on reflection positivity and were thus limited to random surfaces defined on highly symmetric graphs, whereas our argument applies to general graphs. Our proof makes use of a cluster algorithm and reflection transformation for random surfaces of the type introduced by Swendsen-Wang, Wolff and Evertz et al. We discuss the general framework for such cluster algorithms, reviewing several particular cases with emphasis on their use in obtaining theoretical results. Two additional applications are presented: A reflection principle for random surfaces and a proof that pair correlations in the spin $O(n)$ model have monotone densities, strengthening Griffiths' first inequality for such correlations.