Manifold Markov chain Monte Carlo methods for Bayesian inference in diffusion models

Manifold Markov chain Monte Carlo methods for Bayesian inference in diffusion models
复制标题

用于扩散模型中贝叶斯推理的流形马尔可夫链蒙特卡罗方法

DOI:
--
复制
发表时间:
2019
期刊:
Journal of the Royal Statistical Society: Series B (Statistical Methodology)
影响因子:
--
通讯作者:
A. Beskos
A. Beskos
中科院分区:
--
文献类型:
--
作者:
Matthew M. Graham;Alexandre Hoang Thiery;A. Beskos

文献摘要

参考文献

被引文献

相似文献

在离散时间观察到的非线性扩散的贝叶斯推断是一项具有挑战性的任务,它促使了许多算法的发展,主要是在计算统计社区内。我们提出了一个新的方向和相应的方法——借用统计物理和计算化学的思想——来推断潜在扩散路径和模型参数的后验分布,给出了对这一过程的观察。底层过程噪声和参数的联合配置,映射到与观测一致的扩散路径上,形成隐式定义的流形。然后,通过在嵌入流形上使用约束哈密顿蒙特卡罗算法,我们能够对一类离散观察的扩散模型进行计算效率的推断。关键的是,与文献中提出的其他方法相比,我们的方法是高度自动化的,需要最少的用户干预,并适用于一系列设置,包括:椭圆或准椭圆系统;有或没有噪音的观察;线性或非线性观测算子。利用马尔可夫性,我们提出了该方法的一种变体,其复杂性在路径离散化的分辨率和观测次数方面呈线性缩放。可从http://doi.org/10.5281/zenodo.5796148获得重现结果的Python代码。
Bayesian inference for nonlinear diffusions, observed at discrete times, is a challenging task that has prompted the development of a number of algorithms, mainly within the computational statistics community. We propose a new direction, and accompanying methodology—borrowing ideas from statistical physics and computational chemistry—for inferring the posterior distribution of latent diffusion paths and model parameters, given observations of the process. Joint configurations of the underlying process noise and of parameters, mapping onto diffusion paths consistent with observations, form an implicitly defined manifold. Then, by making use of a constrained Hamiltonian Monte Carlo algorithm on the embedded manifold, we are able to perform computationally efficient inference for a class of discretely observed diffusion models. Critically, in contrast with other approaches proposed in the literature, our methodology is highly automated, requiring minimal user intervention and applying alike in a range of settings, including: elliptic or hypo‐elliptic systems; observations with or without noise; linear or non‐linear observation operators. Exploiting Markovianity, we propose a variant of the method with complexity that scales linearly in the resolution of path discretisation and the number of observation times. Python code reproducing the results is available at http://doi.org/10.5281/zenodo.5796148.
DOI: 10.1080/10618600.2013.783484
发表时间: 2013
影响因子: 2.4
作者:
Papaspiliopoulos O
通讯作者: Papaspiliopoulos O
DOI: 10.1016/j.spa.2011.06.003
发表时间: 2011-10-01
影响因子: 1.4
作者:
Beskos, A.;Pinski, F. J.;Stuart, A. M.
通讯作者: Stuart, A. M.
DOI: 10.18637/jss.v076.i01
发表时间: 2017-01-01
影响因子: 5.8
作者:
Carpenter, Bob;Gelman, Andrew;Riddell, Allen
通讯作者: Riddell, Allen
DOI: 10.3150/12-bej414
发表时间: 2013-11-01
期刊: BERNOULLI
影响因子: 1.5
作者:
Beskos, Alexandros;Pillai, Natesh;Stuart, Andrew
通讯作者: Stuart, Andrew
DOI: 10.1016/j.jcp.2016.12.041
发表时间: 2017-04-15
影响因子: 4.1
作者:
Beskos, Alexandros;Girolami, Mark;Stuart, Andrew M.
通讯作者: Stuart, Andrew M.