中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 力学研究所
  3. 中国科学院力学研究所
  4. 国家微重力实验室
Unlocking New Paths for Efficient Analysis of Gravitational Waves from Extreme-Mass-Ratio Inspirals with Machine Learning

文献类型：期刊论文

作者Liang, Bo7,8,9; Guo, Hong6; Zhao TY(赵天宇)9; Wang, He5,7; Evangelinelis, Herik4; Xu YX(许宇翔)7,8,9; Liu, Chang3,9; Liang, Manjia9; Wei XT(魏晓通)9; Yuan, Yong9
刊名CHINESE PHYSICS LETTERS
出版日期2025-08-01
卷号42期号:8页码:9
ISSN号0256-307X
DOI10.1088/0256-307X/42/8/081101
通讯作者Du, Minghui(duminghui@imech.ac.cn) ; Xu, Peng(xupeng@imech.ac.cn)
英文摘要Extreme-mass-ratio inspiral (EMRI) signals pose significant challenges to gravitational wave (GW) data analysis, mainly owing to their highly complex waveforms and high-dimensional parameter space. Given their extended timescales of months to years and low signal-to-noise ratios, detecting and analyzing EMRIs with confidence generally relies on long-term observations. Besides the length of data, parameter estimation is particularly challenging due to non-local parameter degeneracies, arising from multiple local maxima, as well as flat regions and ridges inherent in the likelihood function. These factors lead to exceptionally high time complexity for parameter analysis based on traditional matched filtering and random sampling methods. To address these challenges, the present study explores a machine learning approach to Bayesian posterior estimation of EMRI signals, leveraging the recently developed flow matching technique based on ordinary differential equation neural networks. To our knowledge, this is also the first instance of applying continuous normalizing flows to EMRI analysis. Our approach demonstrates an increase in computational efficiency by several orders of magnitude compared to the traditional Markov chain Monte Carlo (MCMC) methods, while preserving the unbiasedness of results. However, we note that the posterior distributions generated by FMPE may exhibit broader uncertainty ranges than those obtained through full Bayesian sampling, requiring subsequent refinement via methods such as MCMC. Notably, when searching from large priors, our model rapidly approaches the true values while MCMC struggles to converge to the global maximum. Our findings highlight that machine learning has the potential to efficiently handle the vast EMRI parameter space of up to seventeen dimensions, offering new perspectives for advancing space-based GW detection and GW astronomy.
分类号二类/Q1
WOS关键词SPACE ; LISA
资助项目National Key Research and Development Program of China[2021YFC2201901] ; National Key Research and Development Program of China[2021YFC2203004] ; National Key Research and Development Program of China[2020YFC2200100] ; National Key Research and Development Program of China[2021YFC2201903] ; International Partnership Program of the Chinese Academy of Sciences[025GJHZ2023106GC] ; Brazilian agency Fundacao de Amparo a Pesquisa do Estado de So Paulo (FAPESP) ; Brazilian agency Fundacao de Amparo a Pesquisa do Estado do Rio Grande do Sul (FAPERGS) ; Brazilian agency Fundacao de Amparo a Pesquisa do Estado do Rio de Janeiro (FAPERJ) ; Brazilian agency Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) ; Brazilian agency Coordenacao de Aperfeicoamento de Pessoal de Nivel Superior (CAPES)
WOS研究方向Physics
语种英语
WOS记录号WOS:001555818900001
资助机构National Key Research and Development Program of China ; International Partnership Program of the Chinese Academy of Sciences ; Brazilian agency Fundacao de Amparo a Pesquisa do Estado de So Paulo (FAPESP) ; Brazilian agency Fundacao de Amparo a Pesquisa do Estado do Rio Grande do Sul (FAPERGS) ; Brazilian agency Fundacao de Amparo a Pesquisa do Estado do Rio de Janeiro (FAPERJ) ; Brazilian agency Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (CNPq) ; Brazilian agency Coordenacao de Aperfeicoamento de Pessoal de Nivel Superior (CAPES)
其他责任者杜明辉 ; 徐鹏
源URL[http://dspace.imech.ac.cn/handle/311007/103813]  
专题力学研究所_国家微重力实验室
作者单位1.Lanzhou Univ, Lanzhou Ctr Theoret Phys, Lanzhou 730000, Peoples R China
2.Univ Chinese Acad Sci, Hangzhou Inst Adv Study, Key Lab Gravitat Wave Precis Measurement Zhejiang, Hangzhou 310024, Peoples R China;
3.Chinese Acad Sci, Natl Space Sci Ctr, Beijing 100190, Peoples R China;
4.Univ Fed Fluminense, Escola Engn, BR-24210240 Niteroi, Brazil;
5.Univ Chinese Acad Sci, Int Ctr Theoret Phys Asia Pacific, Beijing 100049, Peoples R China;
6.Univ Sao Paulo, Escola Engn Lorena, BR-12602810 Lorena, Brazil;
7.Univ Chinese Acad Sci, Taiji Lab Gravitat Wave Universe Beijing Hangzhou, Beijing 100049, Peoples R China;
8.Chinese Acad Sci, Shanghai Inst Opt & Fine Mech, Shanghai 201800, Peoples R China;
9.Chinese Acad Sci, Inst Mech, Ctr Gravitat Wave Expt, Natl Micrograv Lab, Beijing 100190, Peoples R China;
推荐引用方式
GB/T 7714		 Liang, Bo,Guo, Hong,Zhao TY,et al. Unlocking New Paths for Efficient Analysis of Gravitational Waves from Extreme-Mass-Ratio Inspirals with Machine Learning[J]. CHINESE PHYSICS LETTERS,2025,42(8):9.
APA Liang, Bo.,Guo, Hong.,赵天宇.,Wang, He.,Evangelinelis, Herik.,...&罗子人.(2025).Unlocking New Paths for Efficient Analysis of Gravitational Waves from Extreme-Mass-Ratio Inspirals with Machine Learning.CHINESE PHYSICS LETTERS,42(8),9.
MLA Liang, Bo,et al."Unlocking New Paths for Efficient Analysis of Gravitational Waves from Extreme-Mass-Ratio Inspirals with Machine Learning".CHINESE PHYSICS LETTERS 42.8(2025):9.

入库方式： OAI收割

来源：力学研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。