Birth-death dynamics for sampling: Global convergence, approximations and their asymptotics

Abstract

Motivated by the challenge of sampling Gibbs measures with nonconvex potentials, we study a continuum birth-death dynamics. We improve results in previous works [51,57] and provide weaker hypotheses under which the probability density of the birth-death governed by Kullback-Leibler divergence or by ${\chi }^2$ divergence converge exponentially fast to the Gibbs equilibrium measure, with a universal rate that is independent of the potential barrier. To build a practical numerical sampler based on the pure birth-death dynamics, we consider an interacting particle system, which is inspired by the gradient flow structure and the classical Fokker-Planck equation and relies on kernel-based approximations of the measure. Using the technique of $\Gamma$-convergence of gradient flows, we show that on the torus, smooth and bounded positive solutions of the kernelized dynamics converge on finite time intervals, to the pure birth-death dynamics as the kernel bandwidth shrinks to zero. Moreover we provide quantitative estimates on the bias of minimizers of the energy corresponding to the kernelized dynamics. Finally we prove the long-time asymptotic results on the convergence of the asymptotic states of the kernelized dynamics towards the Gibbs measure.

arXiv commentminor revision at Nonlinearity

References in corpus (21)

• emcee: The MCMC Hammer
• Escaping free-energy minima
• An efficient, multiple range random walk algorithm to calculate the density of states
• Simulated Tempering: A New Monte Carlo Scheme
• Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm
• Optimal Entropy-Transport problems and a new Hellinger-Kantorovich distance between positive measures
• A consensus-based model for global optimization and its mean-field limit
• The Zig-Zag Process and Super-Efficient Sampling for Bayesian Analysis of Big Data
• Randomized Hamiltonian Monte Carlo
• Regularization Matters: Generalization and Optimization of Neural Nets v.s. their Induced Kernel
• Maximum Mean Discrepancy Gradient Flow
• On the geometry of Stein variational gradient descent
• Langevin Monte Carlo and JKO splitting
• Accelerating Langevin Sampling with Birth-death
• On explicit $L^2$-convergence rate estimate for piecewise deterministic Markov processes in MCMC algorithms
• Variational inference via Wasserstein gradient flows
• Polyak-Łojasiewicz inequality on the space of measures and convergence of mean-field birth-death processes
• Kernel Stein Discrepancy Descent
• An Explicit Expansion of the Kullback-Leibler Divergence along its Fisher-Rao Gradient Flow
• Second Order Ensemble Langevin Method for Sampling and Inverse Problems
• Gradient Flows for Sampling: Mean-Field Models, Gaussian Approximations and Affine Invariance

Cited by in corpus (1)

• A connection between Tempering and Entropic Mirror Descent