arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2211.01746stat.COstat.MEstat.ML
收藏

Riemann流形Monte Carlo方法的对数密度梯度协方差与自动度量张量

Log-density gradient covariance and automatic metric tensors for Riemann manifold Monte Carlo methods

Tore Selland Kleppe

AI总结:

本文提出一种基于对数密度梯度协方差(LGC)矩阵的Riemann流形Monte Carlo度量张量,适用于非线性贝叶斯分层模型,能够从复杂先验/潜变量结构自动构建度量张量,结合数值广义随机化Hamiltonian Monte Carlo过程在挑战性目标分布上表现优异。

摘要:

本文提出了一种特别适用于非线性贝叶斯分层模型的Riemann流形Monte Carlo度量张量。该度量张量由对称半正定的对数密度梯度协方差(LGC)矩阵构建,这些矩阵也是本文提出并进一步探讨的。LGC通过测量随机变量及其参数的联合信息内容和依赖结构,推广了Fisher信息矩阵。因此,正定的基于Fisher/LGC的度量张量不仅可以像当前做法那样从观测似然构建,还可以从任意复杂的非线性先验/潜变量结构构建,只要可以为用于构建这些结构的每个条件分布推导出LGC。所提出的方法高度自动化,并允许利用相关模型关联的任何稀疏性。当与最近提出的数值广义随机化Hamiltonian Monte Carlo过程的Riemann流形变体结合实现时,所提出的方法极具竞争力,特别是对于贝叶斯分层模型关联的更具挑战性的目标分布。

英文摘要:

A metric tensor for Riemann manifold Monte Carlo particularly suited for non-linear Bayesian hierarchical models is proposed. The metric tensor is built from symmetric positive semidefinite log-density gradient covariance (LGC) matrices, which are also proposed and further explored here. The LGCs generalize the Fisher information matrix by measuring the joint information content and dependence structure of both a random variable and the parameters of said variable. Consequently, positive definite Fisher/LGC-based metric tensors may be constructed not only from the observation likelihoods as is current practice, but also from arbitrarily complicated non-linear prior/latent variable structures, provided the LGC may be derived for each conditional distribution used to construct said structures. The proposed methodology is highly automatic and allows for exploitation of any sparsity associated with the model in question. When implemented in conjunction with a Riemann manifold variant of the recently proposed numerical generalized randomized Hamiltonian Monte Carlo processes, the proposed methodology is highly competitive, in particular for the more challenging target distributions associated with Bayesian hierarchical models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑