Quantitative low-temperature spectral asymptotics for reversible diffusions in temperature-dependent domains
Noé Blassel, Tony Lelièvre, Gabriel Stoltz
TL;DR
This work studies low-temperature spectral behavior of overdamped Langevin dynamics in domains whose boundaries depend on temperature, revealing how boundary proximity to energy saddles reshapes metastable timescales. By embedding the problem in the Witten Laplacian framework and constructing a harmonic approximation across critical points, the authors prove a leading-order harmonic spectrum limit and a modified Eyring–Kramers formula for the principal eigenvalue, explicitly incorporating boundary-distance parameters via a Gaussian-CDF factor. The analysis introduces geometric boundary-assumptions that localize boundary effects near critical points and develops a Laplace method for moving domains to handle boundary evolution. Practically, the results enable quantitative estimates of exit and decorrelation times for metastable diffusion and inform parameter-tuning in accelerated molecular dynamics methods like ParRep, with explicit prefactors depending on the boundary geometry and Hessian data at critical points.
Abstract
We derive novel low-temperature asymptotics for the spectrum of the infinitesimal generator of the overdamped Langevin dynamics. The novelty is that this operator is endowed with homogeneous Dirichlet conditions at the boundary of a domain which depends on the temperature. From the point of view of stochastic processes, this gives information on the long-time behavior of the diffusion conditioned on non-absorption at the boundary, in the so-called quasistationary regime. Our results provide precise estimates of the spectral gap and principal eigenvalue, extending the Eyring-Kramers formula. The phenomenology is richer than in the case of a fixed boundary and gives new insight into the sensitivity of the spectrum with respect to the shape of the domain near critical points of the energy function. Our work is motivated by--and is relevant to--the problem of finding optimal hyperparameters for accelerated molecular dynamics algorithms.
