arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2206.03369stat.MLcs.LGeess.SPstat.CO
收藏

离散观测扩散过程在线滤波的计算 Doob h-变换

Computational Doob's h-transforms for Online Filtering of Discretely Observed Diffusions

Institut Polytechnique de Paris(巴黎理工学院);ENSAE Paris(巴黎国立统计与经济管理学院);ESSEC Business School(埃塞克商学院);National University of Singapore(新加坡国立大学)

Nicolas Chopin, Andras Fulop, Jeremy Heng, Alexandre H. Thiery

AI总结:

本文针对离散观测非线性扩散过程的在线滤波问题,提出利用神经网络求解倒向 Kolmogorov 方程来近似 Doob h-变换的计算框架,从而训练局部最优辅助粒子滤波器,在高信息量观测、极端观测及高维状态场景下显著提升滤波效率。

摘要:

本文研究离散观测的非线性扩散过程的在线滤波问题。我们的方法基于完全自适应辅助粒子滤波器,其中涉及通常难以解析处理的 Doob h-变换。我们提出一个计算框架,通过使用非线性 Feynman-Kac 公式和神经网络求解底层倒向 Kolmogorov 方程来近似这些 h-变换。该方法允许在数据同化过程之前训练一个局部最优粒子滤波器。数值实验表明,在观测信息高度丰富、观测在模型下属于极端情形、或状态维数较大的情况下,所提出的方法在效率上可比当前最先进的粒子滤波器高出数个数量级。

英文摘要:

This paper is concerned with online filtering of discretely observed nonlinear diffusion processes. Our approach is based on the fully adapted auxiliary particle filter, which involves Doob's $h$-transforms that are typically intractable. We propose a computational framework to approximate these $h$-transforms by solving the underlying backward Kolmogorov equations using nonlinear Feynman-Kac formulas and neural networks. The methodology allows one to train a locally optimal particle filter prior to the data-assimilation procedure. Numerical experiments illustrate that the proposed approach can be orders of magnitude more efficient than state-of-the-art particle filters in the regime of highly informative observations, when the observations are extreme under the model, or if the state dimension is large.

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。

↑