2024-09-17 06:45:43
Ordinary matter is the stuff that makes up everything we know, including astronomical objects that can be directly or indirectly detected by their electromagnetic radiation.
Dark matter is an invisible force that holds the universe together, or so it is believed. By mass, it makes up 85% of all matter, but since we can’t see it directly since it doesn’t emit electromagnetic radiation, we have to study its gravitational effects on galaxies and other cosmic structures. Despite many years of research, the true nature of dark matter (which does not correspond to black holes) remains one of the most enigmatic questions in astrophysics.
According to one dominant theory, dark matter could be a population of particles of a kind that hardly interact with its surroundings except through gravity. But some scientists believe that these particles may occasionally interact with each other, a phenomenon known as self-interaction. The discovery of such interactions would provide crucial clues about the properties of dark matter.
However, the subtle signatures of dark matter’s self-interactions have been difficult to distinguish from other cosmic effects, such as those caused by ordinary matter from active galactic nuclei (the supermassive black holes and their envelopes at the center of galaxies). . Active galactic nuclei can generate effects similar to those of this theoretical interaction of dark matter particles with normal matter, making it difficult to distinguish the latter effects from the aftereffect.
As a significant step forward, astronomer David Harvey, from the Astrophysics Laboratory at the Federal Polytechnic School of Lausanne (EPFL) in Switzerland, has developed an artificial intelligence system that can distinguish between the two types of signals. This new method is designed to distinguish between the effects of self-interactions of dark matter and those caused by active galactic nuclei with normal matter, by analyzing images of galaxy clusters (large groups of galaxies that stay grouped together by the gravity).
This three-dimensional map, obtained thanks to observations made by the Hubble and XMM-Newton space observatories, outlines the large-scale distribution of dark matter in the universe. (Image: NASA/ESA/R. Massey/California Institute of Technology)
Harvey trained a convolutional neural network (a type of artificial intelligence that is very good at recognizing patterns in images) with images from the BAHAMAS-SIDM project, which makes digital models of galaxy clusters under various theoretical scenarios of dark matter and active galactic nuclei. When fed thousands of simulated images of galaxy clusters, the new artificial intelligence learned to distinguish between signals due to self-interactions of dark matter and those due to the effects of active galactic nuclei with normal matter, with an 80 percent success rate .
The innovation promises to greatly expand the scope of dark matter research.
Strategies based on artificial intelligence could be decisive in solving what dark matter really is. As new telescopes specializing in tracking dark matter collect unprecedented amounts of data, the new method will help scientists sift through it quickly and accurately, lending key support to efforts to decipher the nature of the figure out dark matters.
The title of the study is “A deep learning algorithm to disentangle self-interacting dark matter and AGN feedback models.” And it has been published in the academic journal Nature Astronomy. (Source: NYT of Amazings)
#Artificial #intelligence #distinguish #normal #dark #matter