Improved Langevin Monte Carlo for stochastic optimization via landscape modification

by   Michael C. H. Choi, et al.

Given a target function H to minimize or a target Gibbs distribution π_β^0 ∝ e^-β H to sample from in the low temperature, in this paper we propose and analyze Langevin Monte Carlo (LMC) algorithms that run on an alternative landscape as specified by H^f_β,c,1 and target a modified Gibbs distribution π^f_β,c,1∝ e^-β H^f_β,c,1, where the landscape of H^f_β,c,1 is a transformed version of that of H which depends on the parameters f,β and c. While the original Log-Sobolev constant affiliated with π^0_β exhibits exponential dependence on both β and the energy barrier M in the low temperature regime, with appropriate tuning of these parameters and subject to assumptions on H, we prove that the energy barrier of the transformed landscape is reduced which consequently leads to polynomial dependence on both β and M in the modified Log-Sobolev constant associated with π^f_β,c,1. This yield improved total variation mixing time bounds and improved convergence toward a global minimum of H. We stress that the technique developed in this paper is not only limited to LMC and is broadly applicable to other gradient-based optimization or sampling algorithms.


page 1

page 2

page 3

page 4


Landscape modification meets spin systems: from torpid to rapid mixing, tunneling and annealing in the low-temperature regime

Given a target Gibbs distribution π^0_β∝ e^-βℋ to sample from in the low...

On the convergence of an improved discrete simulated annealing via landscape modification

In this paper, we propose new Metropolis-Hastings and simulated annealin...

Logsmooth Gradient Concentration and Tighter Runtimes for Metropolized Hamiltonian Monte Carlo

We show that the gradient norm ∇ f(x) for x ∼(-f(x)), where f is strongl...

Improved Bound for Mixing Time of Parallel Tempering

In the field of sampling algorithms, MCMC (Markov Chain Monte Carlo) met...

Stochastic Approximation Hamiltonian Monte Carlo

Recently, the Hamilton Monte Carlo (HMC) has become widespread as one of...

A simple geometric method for navigating the energy landscape of centroidal Voronoi tessellations

Finding optimal centroidal Voronoi tessellations (CVTs) of a 2D domain p...

Efficient yield optimization with limited gradient information

In this work an efficient strategy for yield optimization with uncertain...

Please sign up or login with your details

Forgot password? Click here to reset