English

Superconcentration in surface growth

Probability 2022-05-10 v3 Statistical Mechanics Mathematical Physics math.MP

Abstract

Height functions of growing random surfaces are often conjectured to be superconcentrated, meaning that their variances grow sublinearly in time. This article introduces a new concept, called subroughness, meaning that there exist two distinct points such that the expected squared difference between the heights at these points grows sublinearly in time. The main result of the paper is that superconcentration is equivalent to subroughness in a class of growing random surfaces. The result is applied to establish superconcentration in a variant of the restricted solid-on-solid (RSOS) model and in a variant of the ballistic deposition model, and give new proofs of superconcentration in directed last-passage percolation and directed polymers.

Keywords

Cite

@article{arxiv.2103.09199,
  title  = {Superconcentration in surface growth},
  author = {Sourav Chatterjee},
  journal= {arXiv preprint arXiv:2103.09199},
  year   = {2022}
}

Comments

32 pages. Minor edits in this revision. To appear in Random Structures and Algorithms

R2 v1 2026-06-24T00:14:44.839Z