Penalized Langevin dynamics with vanishing penalty for smooth and log-concave targets

Abstract

We study the problem of sampling from a probability distribution on ${\mathbb{R}}^p$ defined via a convex and smooth potential function. We consider a continuous-time diffusion-type process, termed Penalized Langevin dynamics (PLD), the drift of which is the negative gradient of the potential plus a linear penalty that vanishes when time goes to infinity. An upper bound on the Wasserstein-2 distance between the distribution of the PLD at time $t$ and the target is established. This upper bound highlights the influence of the speed of decay of the penalty on the accuracy of the approximation. As a consequence, considering the low-temperature limit we infer a new nonasymptotic guarantee of convergence of the penalized gradient flow for the optimization problem.

References in corpus (12)

• A Differential Equation for Modeling Nesterov's Accelerated Gradient Method: Theory and Insights
• A Lyapunov Analysis of Momentum Methods in Optimization
• Convergence of Langevin MCMC in KL-divergence
• Transport Inequalities. A Survey
• Rapid Convergence of the Unadjusted Langevin Algorithm: Isoperimetry Suffices
• The Randomized Midpoint Method for Log-Concave Sampling
• Bounding the error of discretized Langevin algorithms for non-strongly log-concave targets
• Exponential ergodicity of mirror-Langevin diffusions
• An Efficient Sampling Algorithm for Non-smooth Composite Potentials
• Langevin Monte Carlo without smoothness
• Exponential ergodicity for SDEs and McKean-Vlasov processes with Lévy noise
• Statistical Estimation of the Poincar{é} constant and Application to Sampling Multimodal Distributions

Cited by in corpus (1)

• Convergence of Stein Variational Gradient Descent under a Weaker Smoothness Condition