扭转粒子滤波器的指导:连续时间视角

Jianfeng Lu, Yuliang Wang
{"title":"扭转粒子滤波器的指导:连续时间视角","authors":"Jianfeng Lu, Yuliang Wang","doi":"arxiv-2409.02399","DOIUrl":null,"url":null,"abstract":"The particle filter (PF), also known as the sequential Monte Carlo (SMC), is\ndesigned to approximate high-dimensional probability distributions and their\nnormalizing constants in the discrete-time setting. To reduce the variance of\nthe Monte Carlo approximation, several twisted particle filters (TPF) have been\nproposed by researchers, where one chooses or learns a twisting function that\nmodifies the Markov transition kernel. In this paper, we study the TPF from a\ncontinuous-time perspective. Under suitable settings, we show that the\ndiscrete-time model converges to a continuous-time limit, which can be solved\nthrough a series of well-studied control-based importance sampling algorithms.\nThis discrete-continuous connection allows the design of new TPF algorithms\ninspired by established continuous-time algorithms. As a concrete example,\nguided by existing importance sampling algorithms in the continuous-time\nsetting, we propose a novel algorithm called ``Twisted-Path Particle Filter\"\n(TPPF), where the twist function, parameterized by neural networks, minimizes\nspecific KL-divergence between path measures. Some numerical experiments are\ngiven to illustrate the capability of the proposed algorithm.","PeriodicalId":501215,"journal":{"name":"arXiv - STAT - Computation","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Guidance for twisted particle filter: a continuous-time perspective\",\"authors\":\"Jianfeng Lu, Yuliang Wang\",\"doi\":\"arxiv-2409.02399\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The particle filter (PF), also known as the sequential Monte Carlo (SMC), is\\ndesigned to approximate high-dimensional probability distributions and their\\nnormalizing constants in the discrete-time setting. To reduce the variance of\\nthe Monte Carlo approximation, several twisted particle filters (TPF) have been\\nproposed by researchers, where one chooses or learns a twisting function that\\nmodifies the Markov transition kernel. In this paper, we study the TPF from a\\ncontinuous-time perspective. Under suitable settings, we show that the\\ndiscrete-time model converges to a continuous-time limit, which can be solved\\nthrough a series of well-studied control-based importance sampling algorithms.\\nThis discrete-continuous connection allows the design of new TPF algorithms\\ninspired by established continuous-time algorithms. As a concrete example,\\nguided by existing importance sampling algorithms in the continuous-time\\nsetting, we propose a novel algorithm called ``Twisted-Path Particle Filter\\\"\\n(TPPF), where the twist function, parameterized by neural networks, minimizes\\nspecific KL-divergence between path measures. Some numerical experiments are\\ngiven to illustrate the capability of the proposed algorithm.\",\"PeriodicalId\":501215,\"journal\":{\"name\":\"arXiv - STAT - Computation\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-04\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - STAT - Computation\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.02399\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - STAT - Computation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.02399","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

粒子滤波器(PF),又称序列蒙特卡罗(SMC),设计用于在离散时间环境中近似高维概率分布及其归一化常数。为了降低蒙特卡罗近似的方差,研究人员提出了几种扭曲粒子滤波器(TPF),即选择或学习一个扭曲函数来修改马尔科夫转换核。本文从连续时间的角度研究了 TPF。在合适的设置下,我们证明离散时间模型会收敛到连续时间极限,而连续时间极限可以通过一系列经过充分研究的基于控制的重要性采样算法来求解。这种离散-连续的联系使得我们可以从已有的连续时间算法中汲取灵感,设计出新的 TPF 算法。作为一个具体的例子,在连续时间设置中现有重要性采样算法的指导下,我们提出了一种称为 "扭曲路径粒子滤波器"(TPPF)的新算法,其中扭曲函数由神经网络参数化,最小化路径度量之间的特定 KL-发散。本文给出了一些数值实验来说明所提算法的能力。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Guidance for twisted particle filter: a continuous-time perspective
The particle filter (PF), also known as the sequential Monte Carlo (SMC), is designed to approximate high-dimensional probability distributions and their normalizing constants in the discrete-time setting. To reduce the variance of the Monte Carlo approximation, several twisted particle filters (TPF) have been proposed by researchers, where one chooses or learns a twisting function that modifies the Markov transition kernel. In this paper, we study the TPF from a continuous-time perspective. Under suitable settings, we show that the discrete-time model converges to a continuous-time limit, which can be solved through a series of well-studied control-based importance sampling algorithms. This discrete-continuous connection allows the design of new TPF algorithms inspired by established continuous-time algorithms. As a concrete example, guided by existing importance sampling algorithms in the continuous-time setting, we propose a novel algorithm called ``Twisted-Path Particle Filter" (TPPF), where the twist function, parameterized by neural networks, minimizes specific KL-divergence between path measures. Some numerical experiments are given to illustrate the capability of the proposed algorithm.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Statistical Finite Elements via Interacting Particle Langevin Dynamics Graph sub-sampling for divide-and-conquer algorithms in large networks Optimizing VarLiNGAM for Scalable and Efficient Time Series Causal Discovery Best Linear Unbiased Estimate from Privatized Histograms A Bayesian Optimization through Sequential Monte Carlo and Statistical Physics-Inspired Techniques
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1