Product Design, Manufacturing & Innovation Resources
家 » 马尔可夫链蒙特卡罗(MCMC)

马尔可夫链蒙特卡罗(MCMC)

1953
  • Nicholas Metropolis
  • Arianna W. Rosenbluth
  • Marshall N. Rosenbluth
  • Augusta H. Teller
  • Edward Teller
  • W. Keith Hastings
Researcher analyzing Markov Chain Monte Carlo simulations in a statistical analysis office.

(图片仅供参考)

马尔可夫链 Monte Carlo 马尔可夫链蒙特卡罗 (MCMC) 方法是一类用于从概率分布中采样的算法。它构建一个以目标分布为平衡分布或平稳分布的马尔可夫链。经过大量步骤后,链的状态被用作目标分布的样本,从而可以计算积分和期望值。

当直接从复杂的高维概率分布 P(x) 中采样变得困难时,MCMC 方法至关重要。MCMC 不生成独立样本,而是生成一系列相关的样本,这些样本构成马尔可夫链。马尔可夫链是一种随机过程,其中转移到下一状态的概率仅取决于当前状态,而与之前的事件序列无关。关键在于设计链的转移概率,使其平稳分布等于目标分布 P(x)。

该过程从任意状态 x_0 开始。在每个步骤 t,基于当前状态 x_t,使用特定算法(例如 大都会-黑斯廷斯). After an initial “burn-in” period, during which the chain converges from its starting point to the high-probability regions of the target distribution, the subsequent states [latex]x_t, x_{t+1}, …[/latex] can be considered as (correlated) samples from [latex]P(x)[/latex]. These samples can then be used to estimate expectations of functions [latex]f(x)[/latex] with respect to [latex]P(x)[/latex] by averaging [latex]f(x_t)[/latex] over the samples. This is particularly useful in 贝叶斯推断, where [latex]P(x)[/latex] is a posterior distribution of model parameters, and direct calculation is often impossible due to a complex denominator (the evidence or marginal likelihood).

而且: MCMC differs from the basic Monte Carlo method in how it generates samples to estimate a desired distribution or integral. While Monte Carlo methods rely on drawing independent and identically distributed random samples directly from a target distribution or a proposal distribution, MCMC generates samples through a correlated sequence (a Markov chain) where each sample depends on the previous one. This dependency allows MCMC to efficiently explore complex, high-dimensional distributions that are difficult to sample from directly, by constructing a chain that converges to the target distribution over time. In contrast, traditional Monte Carlo methods may struggle with such problems due to inefficiencies in sampling or requiring explicit knowledge of the distribution’s form. Thus, MCMC extends Monte Carlo by harnessing dependence between samples to facilitate sampling in challenging statistical and computational settings.

UNESCO Nomenclature: 1209
– Statistics

类型

Software/Algorithm

中断

Incremental

用法

广泛使用

前体

  • 马尔可夫链理论(安德烈·马尔可夫)
  • foundations of Bayesian statistics (Thomas Bayes, Pierre-Simon Laplace)
  • original 蒙特卡罗方法 (Ulam, Von Neumann)
  • 遍历理论

应用程序

  • 贝叶斯统计用于参数估计
  • 计算生物学在系统发育树推断中的应用
  • 用于训练概率模型的机器学习
  • 计算物理学在模拟分子系统中的应用
  • 计量经济学在复杂金融数据建模中的应用

专利:

NA

潜在创新理念

  • automated probabilistic programming languages that use MCMC as a backend
  • 基于贝叶斯模型的流数据实时异常检测
  • 利用宇宙微波背景辐射数据推断宇宙学参数
  • 重建古代语言和进化树
  • 基于学习到的概率分布的生成式艺术和音乐创作
  • 通过探索分子构象空间进行药物发现
  • 通过对未来可能需求分布进行抽样来优化供应链
  • 基于市场行为概率模型的个性化财务建议
  • natural language processing for topic modeling and sentiment analysis
  • 利用气候模型量化未来气候预测的不确定性
相关术语:MCMC、马尔可夫链、贝叶斯推断、统计学、抽样、平稳分布、Metropolis-Hastings算法、吉布斯抽样、计算统计学、后验分布。

历史背景

马尔可夫链蒙特卡罗(MCMC)

1943
1950
1950
1953
1960
1960
1967
1940
1950
1950
1952
1956
1960
1967
1967

(如果日期未知或不相关,例如“流体力学”,则提供其显著出现的近似估计)

第十年 周年纪念日

只有注册会员才能免费获取完整尺寸的图片和下载内容。

> 登录 <