Sampling with Prior Knowledge for High-dimensional Gravitational Wave Data Analysis
Extracting knowledge from high-dimensional data has been notoriously difficult, primarily due to the so-called "curse of dimensionality" and the complex joint distributions of these dimensions. This is a particularly profound issue for high-dimensional gravitational wave data analysis wher...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2022-03-01
|
| Series: | Big Data Mining and Analytics |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.26599/BDMA.2021.9020018 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832572803357868032 |
|---|---|
| author | He Wang Zhoujian Cao Yue Zhou Zong-Kuan Guo Zhixiang Ren |
| author_facet | He Wang Zhoujian Cao Yue Zhou Zong-Kuan Guo Zhixiang Ren |
| author_sort | He Wang |
| collection | DOAJ |
| description | Extracting knowledge from high-dimensional data has been notoriously difficult, primarily due to the so-called "curse of dimensionality" and the complex joint distributions of these dimensions. This is a particularly profound issue for high-dimensional gravitational wave data analysis where one requires to conduct Bayesian inference and estimate joint posterior distributions. In this study, we incorporate prior physical knowledge by sampling from desired interim distributions to develop the training dataset. Accordingly, the more relevant regions of the high-dimensional feature space are covered by additional data points, such that the model can learn the subtle but important details. We adapt the normalizing flow method to be more expressive and trainable, such that the information can be effectively extracted and represented by the transformation between the prior and target distributions. Once trained, our model only takes approximately 1 s on one V100 GPU to generate thousands of samples for probabilistic inference purposes. The evaluation of our approach confirms the efficacy and efficiency of gravitational wave data inferences and points to a promising direction for similar research. The source code, specifications, and detailed procedures are publicly accessible on GitHub. |
| format | Article |
| id | doaj-art-4da4a65da12c47ebafc8269f80946c39 |
| institution | Kabale University |
| issn | 2096-0654 |
| language | English |
| publishDate | 2022-03-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Big Data Mining and Analytics |
| spelling | doaj-art-4da4a65da12c47ebafc8269f80946c392025-02-02T06:50:34ZengTsinghua University PressBig Data Mining and Analytics2096-06542022-03-0151536310.26599/BDMA.2021.9020018Sampling with Prior Knowledge for High-dimensional Gravitational Wave Data AnalysisHe Wang0Zhoujian Cao1Yue Zhou2Zong-Kuan Guo3Zhixiang Ren4CAS Key Laboratory of Theoretical Physics, Institute of Theoretical Physics, Chinese Academy of Sciences, P.O. Box 2735, Beijing 100190, ChinaDepartment of Astronomy, Beijing Normal University, Beijing 100875, ChinaDepartment of Networked Intelligence, Peng Cheng Laboratory, Shenzhen 518055, ChinaCAS Key Laboratory of Theoretical Physics, Institute of Theoretical Physics, Chinese Academy of Sciences, P.O. Box 2735, Beijing 100190, ChinaDepartment of Networked Intelligence, Peng Cheng Laboratory, Shenzhen 518055, ChinaExtracting knowledge from high-dimensional data has been notoriously difficult, primarily due to the so-called "curse of dimensionality" and the complex joint distributions of these dimensions. This is a particularly profound issue for high-dimensional gravitational wave data analysis where one requires to conduct Bayesian inference and estimate joint posterior distributions. In this study, we incorporate prior physical knowledge by sampling from desired interim distributions to develop the training dataset. Accordingly, the more relevant regions of the high-dimensional feature space are covered by additional data points, such that the model can learn the subtle but important details. We adapt the normalizing flow method to be more expressive and trainable, such that the information can be effectively extracted and represented by the transformation between the prior and target distributions. Once trained, our model only takes approximately 1 s on one V100 GPU to generate thousands of samples for probabilistic inference purposes. The evaluation of our approach confirms the efficacy and efficiency of gravitational wave data inferences and points to a promising direction for similar research. The source code, specifications, and detailed procedures are publicly accessible on GitHub.https://www.sciopen.com/article/10.26599/BDMA.2021.9020018high-dimensional dataprior samplingnormalizing flowgravitational wave |
| spellingShingle | He Wang Zhoujian Cao Yue Zhou Zong-Kuan Guo Zhixiang Ren Sampling with Prior Knowledge for High-dimensional Gravitational Wave Data Analysis Big Data Mining and Analytics high-dimensional data prior sampling normalizing flow gravitational wave |
| title | Sampling with Prior Knowledge for High-dimensional Gravitational Wave Data Analysis |
| title_full | Sampling with Prior Knowledge for High-dimensional Gravitational Wave Data Analysis |
| title_fullStr | Sampling with Prior Knowledge for High-dimensional Gravitational Wave Data Analysis |
| title_full_unstemmed | Sampling with Prior Knowledge for High-dimensional Gravitational Wave Data Analysis |
| title_short | Sampling with Prior Knowledge for High-dimensional Gravitational Wave Data Analysis |
| title_sort | sampling with prior knowledge for high dimensional gravitational wave data analysis |
| topic | high-dimensional data prior sampling normalizing flow gravitational wave |
| url | https://www.sciopen.com/article/10.26599/BDMA.2021.9020018 |
| work_keys_str_mv | AT hewang samplingwithpriorknowledgeforhighdimensionalgravitationalwavedataanalysis AT zhoujiancao samplingwithpriorknowledgeforhighdimensionalgravitationalwavedataanalysis AT yuezhou samplingwithpriorknowledgeforhighdimensionalgravitationalwavedataanalysis AT zongkuanguo samplingwithpriorknowledgeforhighdimensionalgravitationalwavedataanalysis AT zhixiangren samplingwithpriorknowledgeforhighdimensionalgravitationalwavedataanalysis |