Machine learning Post-Minkowskian integrals
Abstract We study a neural network framework for the numerical evaluation of Feynman loop integrals that are fundamental building blocks for perturbative computations of physical observables in gauge and gravity theories. We show that such a machine learning approach improves the convergence of the...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-07-01
|
| Series: | Journal of High Energy Physics |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/JHEP07(2023)181 |
| _version_ | 1797647471416967168 |
|---|---|
| author | Ryusuke Jinno Gregor Kälin Zhengwen Liu Henrique Rubira |
| author_facet | Ryusuke Jinno Gregor Kälin Zhengwen Liu Henrique Rubira |
| author_sort | Ryusuke Jinno |
| collection | DOAJ |
| description | Abstract We study a neural network framework for the numerical evaluation of Feynman loop integrals that are fundamental building blocks for perturbative computations of physical observables in gauge and gravity theories. We show that such a machine learning approach improves the convergence of the Monte Carlo algorithm for high-precision evaluation of multi-dimensional integrals compared to traditional algorithms. In particular, we use a neural network to improve the importance sampling. For a set of representative integrals appearing in the computation of the conservative dynamics for a compact binary system in General Relativity, we perform a quantitative comparison between the Monte Carlo integrators VEGAS and i-flow, an integrator based on neural network sampling. |
| first_indexed | 2024-03-11T15:18:01Z |
| format | Article |
| id | doaj.art-4ea9c703dd2f431e9e595dacfbf39d53 |
| institution | Directory Open Access Journal |
| issn | 1029-8479 |
| language | English |
| last_indexed | 2024-03-11T15:18:01Z |
| publishDate | 2023-07-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Journal of High Energy Physics |
| spelling | doaj.art-4ea9c703dd2f431e9e595dacfbf39d532023-10-29T12:06:29ZengSpringerOpenJournal of High Energy Physics1029-84792023-07-012023714510.1007/JHEP07(2023)181Machine learning Post-Minkowskian integralsRyusuke Jinno0Gregor Kälin1Zhengwen Liu2Henrique Rubira3Instituto de Física Teórica UAM/CSICDeutsches Elektronen-Synchrotron DESYDeutsches Elektronen-Synchrotron DESYDeutsches Elektronen-Synchrotron DESYAbstract We study a neural network framework for the numerical evaluation of Feynman loop integrals that are fundamental building blocks for perturbative computations of physical observables in gauge and gravity theories. We show that such a machine learning approach improves the convergence of the Monte Carlo algorithm for high-precision evaluation of multi-dimensional integrals compared to traditional algorithms. In particular, we use a neural network to improve the importance sampling. For a set of representative integrals appearing in the computation of the conservative dynamics for a compact binary system in General Relativity, we perform a quantitative comparison between the Monte Carlo integrators VEGAS and i-flow, an integrator based on neural network sampling.https://doi.org/10.1007/JHEP07(2023)181Scattering AmplitudesClassical Theories of GravityEffective Field Theories |
| spellingShingle | Ryusuke Jinno Gregor Kälin Zhengwen Liu Henrique Rubira Machine learning Post-Minkowskian integrals Journal of High Energy Physics Scattering Amplitudes Classical Theories of Gravity Effective Field Theories |
| title | Machine learning Post-Minkowskian integrals |
| title_full | Machine learning Post-Minkowskian integrals |
| title_fullStr | Machine learning Post-Minkowskian integrals |
| title_full_unstemmed | Machine learning Post-Minkowskian integrals |
| title_short | Machine learning Post-Minkowskian integrals |
| title_sort | machine learning post minkowskian integrals |
| topic | Scattering Amplitudes Classical Theories of Gravity Effective Field Theories |
| url | https://doi.org/10.1007/JHEP07(2023)181 |
| work_keys_str_mv | AT ryusukejinno machinelearningpostminkowskianintegrals AT gregorkalin machinelearningpostminkowskianintegrals AT zhengwenliu machinelearningpostminkowskianintegrals AT henriquerubira machinelearningpostminkowskianintegrals |