A scientific team led by the Institute of Astrophysics and Space Sciences (IA) created, together with a company, a machine learning method that recognizes ‘superluminous’ galaxies from the early days of the Universe, it was announced this Wednesday.
The method is described in an article published in the journal Astronomy & Astrophysics and, according to its authors, “will allow astronomers to be more effective in the search for so-called radio galaxies”, galaxies “with an active nucleus that emit powerful jets of matter, that shine at radio frequencies,” the AI says in a statement.
The international team led by AI developed, in collaboration with the company Closer, a specialist in artificial intelligence, an algorithm that was trained with images of galaxies at various wavelengths of light.
When tested with other images, the algorithm proved to be “capable of predicting four times more radio galaxies than conventional methods with explicit instructions,” the AI statement states, adding that machine learning (a branch of artificial intelligence) could ” help clarify physical phenomena.” that were happening” in radio galaxies “when the Universe was a tenth of its current age.
One of the authors of the article, José Afonso, an IA researcher specialized in the study of galaxies, highlighted, cited in the same statement, the importance for astronomy of the “development of advanced techniques” for the processing and analysis of “enormous amounts of data”.
“At AI we are developing and implementing these techniques to be able to decipher the origin of galaxies and the supermassive black holes that many host,” he highlighted.
Source: TSF
