English
Related papers

Related papers: Optimizing the diffusion coefficient of overdamped…

200 papers

A systematic procedure for optimising the friction coefficient in underdamped Langevin dynamics as a sampling tool is given by taking the gradient of the associated asymptotic variance with respect to friction. We give an expression for…

Computation · Statistics 2023-11-01 Martin Chak , Nikolas Kantas , Tony Lelièvre , Grigorios A. Pavliotis

Markov chain Monte Carlo samplers based on discretizations of (overdamped) Langevin dynamics are commonly used in the Bayesian inference and computational statistical physics literature to estimate high-dimensional integrals. One can…

Numerical Analysis · Mathematics 2025-08-11 Tony Lelièvre , Régis Santet , Gabriel Stoltz

In this paper, we provide a multiscale perspective on the problem of maximum marginal likelihood estimation. We consider and analyse a diffusion-based maximum marginal likelihood estimation scheme using ideas from multiscale dynamics. Our…

Computation · Statistics 2024-06-11 O. Deniz Akyildiz , Michela Ottobre , Iain Souttar

In this paper we introduce and analyse Langevin samplers that consist of perturbations of the standard underdamped Langevin dynamics. The perturbed dynamics is such that its invariant measure is the same as that of the unperturbed dynamics.…

Probability · Mathematics 2017-12-06 A. B. Duncan , N. Nuesken , G. A. Pavliotis

In this paper we consider a new probability sampling methods based on Langevin diffusion dynamics to resolve the problem of existing Monte Carlo algorithms when draw samples from high dimensional target densities. We extent…

Machine Learning · Computer Science 2025-03-31 Z. Zarezadeh , N. Zarezadeh

Langevin diffusion processes and their discretizations are often used for sampling from a target density. The most convenient framework for assessing the quality of such a sampling scheme corresponds to smooth and strongly log-concave…

Probability · Mathematics 2018-12-27 Arnak S. Dalalyan , Lionel Riou-Durand

We study the convergence to equilibrium of an underdamped Langevin equation that is controlled by a linear feedback force. Specifically, we are interested in sampling the possibly multimodal invariant probability distribution of a Langevin…

Optimization and Control · Mathematics 2022-01-12 Tobias Breiten , Carsten Hartmann , Lara Neureither , Upanshu Sharma

Adaptive Langevin dynamics is a method for sampling the Boltzmann-Gibbs distribution at prescribed temperature in cases where the potential gradient is subject to stochastic perturbation of unknown magnitude. The method replaces the…

Probability · Mathematics 2023-11-14 Benedict Leimkuhler , Matthias Sachs , Gabriel Stoltz

It is well known that adding any skew symmetric matrix to the gradient of Langevin dynamics algorithm results in a non-reversible diffusion with improved convergence rate. This paper presents a gradient algorithm to adaptively optimize the…

Machine Learning · Computer Science 2020-09-29 Vikram Krishnamurthy , George Yin

We study the exponential convergence to the stationary state for nonequilibrium Langevin dynamics, by a perturbative approach based on hypocoercive techniques developed for equilibrium Langevin dynamics. The Hamiltonian and overdamped…

Mathematical Physics · Physics 2017-07-06 Alessandra Iacobucci , Stefano Olla , Gabriel Stoltz

We construct a novel estimator for the diffusion coefficient of the limiting homogenized equation, when observing the slow dynamics of a multiscale model, in the case when the slow dynamics are of bounded variation. Previous research…

Statistics Theory · Mathematics 2018-07-04 Theodoros Manikas , Anastasia Papavasiliou

Many methods that build powerful variational distributions based on unadjusted Langevin transitions exist. Most of these were developed using a wide range of different approaches and techniques. Unfortunately, the lack of a unified analysis…

Machine Learning · Computer Science 2023-03-24 Tomas Geffner , Justin Domke

Langevin dynamics has become a popular tool to simulate the Boltzmann equilibrium distribution. When the repartition of the Langevin equation involves the exact realization of the Ornstein-Uhlenbeck noise, in addition to the conventional…

Chemical Physics · Physics 2017-11-15 Dezhang Li , Xu Han , Yichen Chai , Cong Wang , Zifei Chen , Zhijun Zhang , Jian Liu , Jiushu Shao

A model has two main aims: predicting the behavior of a physical system and understanding its nature, that is how it works, at some desired level of abstraction. A promising recent approach to model building consists in deriving a…

Statistical Mechanics · Physics 2019-02-26 Marco Baldovin , Andrea Puglisi , Angelo Vulpiani

The probabilistic diffusion model has become highly effective across various domains. Typically, sampling from a diffusion model involves using a denoising distribution characterized by a Gaussian with a learned mean and either fixed or…

Machine Learning · Computer Science 2025-02-20 Zijing Ou , Mingtian Zhang , Andi Zhang , Tim Z. Xiao , Yingzhen Li , David Barber

We consider the setting of multiscale overdamped Langevin stochastic differential equations, and study the problem of learning the drift function of the homogenized dynamics from continuous-time observations of the multiscale system. We…

Numerical Analysis · Mathematics 2024-11-12 Max Hirsch , Andrea Zanoni

This article considers the application of Langevin dynamics to sampling and investigates how to choose the damping parameter in Langevin dynamics for the purpose of maximizing thoroughness of sampling. Also, it considers the computation of…

Computation · Statistics 2021-06-23 Robert D. Skeel , Carsten Hartmann

Drawing from the theory of stochastic differential equations, we introduce a novel sampling method for known distributions and a new algorithm for diffusion generative models with unknown distributions. Our approach is inspired by the…

Statistics Theory · Mathematics 2024-07-12 Xicheng Zhang

The stochastic motion in a nonhomogeneous medium with traps is studied and diffusion properties of that system are discussed. The particle is subjected to a stochastic stimulation obeying a general L\'evy stable statistics and experiences…

Statistical Mechanics · Physics 2015-06-11 Tomasz Srokowski

We explore properties the solution of Langevin equation when stochastic influence is orthogonal to velocity of a particle. Wiener's process can accept unlimited values. But for these equations, the attraction surfaces exist. For these…

Probability · Mathematics 2019-06-20 V. A. Doobko
‹ Prev 1 2 3 10 Next ›