arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1312.4852stat.MLcs.SYeess.SY
收藏

基于粒子随机近似EM算法的高斯过程状态空间模型识别

Identification of Gaussian Process State-Space Models with Particle Stochastic Approximation EM

University of Cambridge(剑桥大学);Linköping University(林雪平大学);Uppsala University(乌普萨拉大学)

Roger Frigola, Fredrik Lindsten, Thomas B. Schön, Carl E. Rasmussen

AI总结:

本文提出基于粒子随机近似EM算法的高斯过程状态空间模型参数识别方法,通过有效利用粒子MCMC技术实现非参数动态描述的系统参数估计。

摘要:

高斯过程状态空间模型(GP-SSMs)是一类非常灵活的非线性动态系统建模方法,包含对系统动态的贝叶斯非参数表示以及调控该表示性质的附加(超)参数。贝叶斯框架使得系统动态的不确定性可以系统性地进行推理。本文提出了一种在保留完整非参数动态描述的同时,对GP-SSMs参数进行最大似然识别的方法,该方法基于随机近似版本的EM算法,并利用最近的粒子马尔可夫链蒙特卡洛发展实现高效识别。

英文摘要:

Gaussian process state-space models (GP-SSMs) are a very flexible family of models of nonlinear dynamical systems. They comprise a Bayesian nonparametric representation of the dynamics of the system and additional (hyper-)parameters governing the properties of this nonparametric representation. The Bayesian formalism enables systematic reasoning about the uncertainty in the system dynamics. We present an approach to maximum likelihood identification of the parameters in GP-SSMs, while retaining the full nonparametric description of the dynamics. The method is based on a stochastic approximation version of the EM algorithm that employs recent developments in particle Markov chain Monte Carlo for efficient identification.

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。

↑