BriefGPT.xyz
Jun, 2013
利用粒子MCMC对高斯过程状态空间模型进行贝叶斯推断和学习
Bayesian Inference and Learning in Gaussian Process State-Space Models with Particle MCMC
HTML
PDF
Roger Frigola, Fredrik Lindsten, Thomas B. Schön, Carl E. Rasmussen
TL;DR
通过向状态转移动力学分布中添加高斯过程先验，结合分析型建模和蒙特卡罗采样器进行直接联合平滑分布推断的方法，提出了一种非线性非参数状态空间模型的完全贝叶斯方法。
Abstract
state-space models
are successfully used in many areas of science, engineering and economics to model time series and dynamical systems. We present a fully Bayesian approach to inference and learning in nonlinear nonparametric
→