English
Related papers

Related papers: Contractive kinetic Langevin samplers beyond globa…

200 papers

The Metropolis-adjusted Langevin algorithm (MALA) is a Metropolis-Hastings method for approximate sampling from continuous distributions. We derive upper bounds for the contraction rate in Kantorovich-Rubinstein-Wasserstein distance of the…

Probability · Mathematics 2014-01-17 Andreas Eberle

We present a unified framework to analyze the global convergence of Langevin dynamics based algorithms for nonconvex finite-sum optimization with $n$ component functions. At the core of our analysis is a direct analysis of the ergodicity of…

Machine Learning · Statistics 2020-10-20 Pan Xu , Jinghui Chen , Difan Zou , Quanquan Gu

We study sampling problems associated with non-convex potentials that meanwhile lack smoothness. In particular, we consider target distributions that satisfy either logarithmic-Sobolev inequality or Poincar\'e inequality. Rather than…

Machine Learning · Computer Science 2023-02-21 Jiaming Liang , Yongxin Chen

We analyze a recently proposed class of algorithms for the problem of sampling from probability distributions $\mu^\ast$ in $\mathbb{R}^d$ with a Lebesgue density of the form $\mu^\ast(x) \propto \exp(-f(Kx)-g(x))$, where $K$ is a linear…

Optimization and Control · Mathematics 2024-11-06 Martin Burger , Matthias J. Ehrhardt , Lorenz Kuger , Lukas Weigand

We study sampling from a target distribution $\nu_* = e^{-f}$ using the unadjusted Langevin Monte Carlo (LMC) algorithm when the potential $f$ satisfies a strong dissipativity condition and it is first-order smooth with a Lipschitz…

Machine Learning · Statistics 2021-07-09 Murat A. Erdogdu , Rasa Hosseinzadeh , Matthew S. Zhang

We study the proximal sampler of Lee, Shen, and Tian (2021) and obtain new convergence guarantees under weaker assumptions than strong log-concavity: namely, our results hold for (1) weakly log-concave targets, and (2) targets satisfying…

Statistics Theory · Mathematics 2022-02-15 Yongxin Chen , Sinho Chewi , Adil Salim , Andre Wibisono

Sampling from probability distributions of the form $\sigma \propto e^{-\beta V}$, where $V$ is a continuous potential, is a fundamental task across physics, chemistry, biology, computer science, and statistics. However, when $V$ is…

Quantum Physics · Physics 2026-02-24 Jiaqi Leng , Zhiyan Ding , Zherui Chen , Lin Lin

In this paper we establish quantitative results about the bridges of the Langevin dynamics and the associated reciprocal processes. They include an equivalence between gradient estimates for bridge semigroups and couplings, comparison…

Probability · Mathematics 2016-12-30 Giovanni Conforti , Max Von Renesse

We propose discrete Langevin proposal (DLP), a simple and scalable gradient-based proposal for sampling complex high-dimensional discrete distributions. In contrast to Gibbs sampling-based methods, DLP is able to update all coordinates in…

Machine Learning · Computer Science 2022-06-22 Ruqi Zhang , Xingchao Liu , Qiang Liu

Stochastic learning dynamics based on Langevin or Levy stochastic differential equations (SDEs) in deep neural networks control the variance of noise by varying the size of the mini-batch or directly those of injecting noise. Since the…

Machine Learning · Computer Science 2023-10-05 JInwuk Seok , Changsik Cho

We study the convergence in total variation and $V$-norm of discretization schemes of the underdamped Langevin dynamics. Such algorithms are very popular and commonly used in molecular dynamics and computational statistics to…

Probability · Mathematics 2023-04-24 Alain Durmus , Aurélien Enfroy , Éric Moulines , Gabriel Stoltz

Score-based Generative Models (SGMs) aim to sample from a target distribution by learning score functions using samples perturbed by Gaussian noise. Existing convergence bounds for SGMs in the W2-distance rely on stringent assumptions about…

Machine Learning · Statistics 2025-12-12 Marta Gentiloni-Silveri , Antonio Ocello

A key task in Bayesian machine learning is sampling from distributions that are only specified up to a partition function (i.e., constant of proportionality). One prevalent example of this is sampling posteriors in parametric distributions,…

Machine Learning · Computer Science 2020-09-10 Rong Ge , Holden Lee , Andrej Risteski

Continuous-time models provide important insights into the training dynamics of optimization algorithms in deep learning. In this work, we establish a non-asymptotic convergence analysis of stochastic gradient Langevin dynamics (SGLD),…

Machine Learning · Computer Science 2026-01-30 Noah Oberweis , Semih Cayci

We consider the problem of selective inference after solving a (randomized) convex statistical learning program in the form of a penalized or constrained loss function. Our first main result is a change-of-measure formula that describes…

Statistics Theory · Mathematics 2016-09-20 Xiaoying Tian Harris , Snigdha Panigrahi , Jelena Markovic , Nan Bi , Jonathan Taylor

We prove quantitative convergence rates at which discrete Langevin-like processes converge to the invariant distribution of a related stochastic differential equation. We study the setup where the additive noise can be non-Gaussian and…

Machine Learning · Computer Science 2020-11-20 Xiang Cheng , Dong Yin , Peter L. Bartlett , Michael I. Jordan

A canonical algorithm for log-concave sampling is the Langevin Algorithm, aka the Langevin Diffusion run with some discretization stepsize $\eta > 0$. This discretization leads the Langevin Algorithm to have a stationary distribution…

Machine Learning · Statistics 2024-10-22 Jason M. Altschuler , Kunal Talwar

Classically, the continuous-time Langevin diffusion converges exponentially fast to its stationary distribution $\pi$ under the sole assumption that $\pi$ satisfies a Poincar\'e inequality. Using this fact to provide guarantees for the…

Statistics Theory · Mathematics 2024-07-11 Sinho Chewi , Murat A. Erdogdu , Mufan Bill Li , Ruoqi Shen , Matthew Zhang

Sampling from a target distribution is a fundamental problem. Traditional Markov chain Monte Carlo (MCMC) algorithms, such as the unadjusted Langevin algorithm (ULA), derived from the overdamped Langevin dynamics, have been extensively…

Optimization and Control · Mathematics 2024-10-29 Xinzhe Zuo , Stanley Osher , Wuchen Li

Coupling arguments are a central tool for bounding the deviation between two stochastic processes, but traditionally have been limited to Wasserstein metrics. In this paper, we apply the shifted composition rule--an information-theoretic…

Statistics Theory · Mathematics 2024-12-25 Jason M. Altschuler , Sinho Chewi