Non-asymptotic estimates for accelerated high order Langevin Monte Carlo algorithms

📅 2024-05-09
📈 Citations: 1
Influential: 0
📄 PDF

career value

244K/year
🤖 AI Summary
Sampling from high-dimensional target distributions with superlinearly growing potentials—such as nonconvex functions that become convex at infinity—remains challenging. Method: We propose two accelerated higher-order Langevin algorithms, aHOLA and aHOLLA, which integrate higher-order discretizations of Langevin dynamics with momentum-based acceleration. Our theoretical analysis leverages local Hölder continuity, convexity at infinity, and dissipativity conditions. Contribution/Results: We establish the first non-asymptotic convergence bound in Wasserstein-1/2 distance for nonconvex settings, achieving a Wasserstein-1 convergence rate of order $1 + q/2$, where $q$ is the Hölder exponent—surpassing existing rates and attaining the best-known rate for nonconvex sampling. Numerical experiments across diverse nonconvex distributions validate the algorithms’ accelerated convergence, stability, and practical efficiency.

Technology Category

Application Category

📝 Abstract
In this paper, we propose two new algorithms, namely aHOLA and aHOLLA, to sample from high-dimensional target distributions with possibly super-linearly growing potentials. We establish non-asymptotic convergence bounds for aHOLA in Wasserstein-1 and Wasserstein-2 distances with rates of convergence equal to $1+q/2$ and $1/2+q/4$, respectively, under a local H""{o}lder condition with exponent $qin(0,1]$ and a convexity at infinity condition on the potential of the target distribution. Similar results are obtained for aHOLLA under certain global continuity conditions and a dissipativity condition. Crucially, we achieve state-of-the-art rates of convergence of the proposed algorithms in the non-convex setting which are higher than those of the existing algorithms. Numerical experiments are conducted to sample from several distributions and the results support our main findings.
Problem

Research questions and friction points this paper is trying to address.

Develop accelerated algorithms for high-dimensional sampling
Establish non-asymptotic convergence bounds for new algorithms
Achieve higher convergence rates in non-convex settings
Innovation

Methods, ideas, or system contributions that make the work stand out.

Accelerated high order Langevin Monte Carlo algorithms
Non-asymptotic convergence bounds in Wasserstein distances
State-of-the-art rates in non-convex settings
🔎 Similar Papers