Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2510.18964 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909863155597312 |
|---|---|
| author | Rios, Martín de los Di Gioia, Serafina Iocco, Fabio Trotta, Roberto |
| author_facet | Rios, Martín de los Di Gioia, Serafina Iocco, Fabio Trotta, Roberto |
| contents | Machine learning has the potential to improve the reconstruction of the dark matter profile of galaxies with respect to traditional methods, like rotation curves. We demonstrate on the simulation suite Illustris-TNG that a steerable equivariant convolutional neural network (CNN) is able to infer the dark matter profiles within and around individual galaxies from photometric and interferometric data, improving on a standard CNN. Within the in silico environment of the simulations, our architecture is able to capture the dark matter distribution within galaxies without a parametrization of the profile. We perform an interpretability analysis to understand the internal mechanisms of the trained model and the most important data features used to estimate the dark matter profiles. The equivariant CNN recovers the dark matter profile of galaxies within the stellar mass range $[10^{10} - 10^{12} ]$ $M_{\odot}$ with excellent precision and accuracy: the mean squared error is reduced by a factor of ~ 3 from its value under the training distribution, demonstrating that the network has learnt from the data features. While this holds within the controlled 'in silico' environment of the simulation, we argue that few additional steps are needed before this method can be reliably applied to galaxies in the real field observations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_18964 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Dark Matter profiles of "in silico" galaxies: deep learning inference Rios, Martín de los Di Gioia, Serafina Iocco, Fabio Trotta, Roberto Astrophysics of Galaxies Data Analysis, Statistics and Probability Machine learning has the potential to improve the reconstruction of the dark matter profile of galaxies with respect to traditional methods, like rotation curves. We demonstrate on the simulation suite Illustris-TNG that a steerable equivariant convolutional neural network (CNN) is able to infer the dark matter profiles within and around individual galaxies from photometric and interferometric data, improving on a standard CNN. Within the in silico environment of the simulations, our architecture is able to capture the dark matter distribution within galaxies without a parametrization of the profile. We perform an interpretability analysis to understand the internal mechanisms of the trained model and the most important data features used to estimate the dark matter profiles. The equivariant CNN recovers the dark matter profile of galaxies within the stellar mass range $[10^{10} - 10^{12} ]$ $M_{\odot}$ with excellent precision and accuracy: the mean squared error is reduced by a factor of ~ 3 from its value under the training distribution, demonstrating that the network has learnt from the data features. While this holds within the controlled 'in silico' environment of the simulation, we argue that few additional steps are needed before this method can be reliably applied to galaxies in the real field observations. |
| title | Dark Matter profiles of "in silico" galaxies: deep learning inference |
| topic | Astrophysics of Galaxies Data Analysis, Statistics and Probability |
| url | https://arxiv.org/abs/2510.18964 |