[Curiosities] Artificial intelligence has allowed us to improve the first image of a black hole

[Draeno]

 

Messier 87 is a huge elliptical galaxy located approximately 53 million light-years from Earth in the constellation Virgo. In April 2019, the EHT (Event Horizon Telescope) collaboration shared amazing images of M87*, the supermassive black hole located at the core of Messier 87. These images were obtained thanks to EHT observations carried out in April 2017.

To collect this information, the EHT astronomers used an array of seven radio telescopes distributed around the world, thus creating a virtual telescope the size of our planet, capable of observing the "shadow" of the event horizon of a black hole thanks to its power and resolution.

Black holes naturally cannot be seen because their gravity does not let light escape; but an active supermassive black hole, or one that is actively feeding on matter, has a hot disk and a glowing torus of material around it.

While this method employed by EHT allowed the team to visualize surprisingly precise details of this mysterious event, there were still many gaps in the information obtained.

Now, in a new study published in the Astrophysical Journal, it has been described how a new artificial intelligence machine learning technique has been used to fill these gaps in the data.

ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
PRIMO is based on a strand of machine learning that teaches an artificial intelligence certain rules by exposing it to thousands of examples. The effectiveness of this type of machine learning has been demonstrated on multiple occasions, such as the culmination of Beethoven's unfinished work.

Using PRIMO on the EHT image of M87*, the system examined more than 30,000 high-quality simulated images of gas accumulating in a black hole to identify recurring patterns in the images.

The results were then merged to provide a highly accurate representation of the EHT observations, while providing a high-quality estimate of the missing structure in the image.

In this way, the researchers thus confirmed that the newly generated image of M87* is consistent with the EHT data and with theoretical expectations, including the bright ring of emission that is expected to be produced by the hot gas being absorbed by the black hole. .

The new image should provide more accurate estimates of the black hole's mass and the physical parameters that influence its current appearance. In addition, PRIMO can be applied to other EHT observations, such as those of Sagittarius A*, the central black hole of our own Milky Way galaxy.

 

https://www.nationalgeographic.com.es/ciencia/inteligencia-artificial-ha-permitido-mejorar-primera-imagen-agujero-negro_19775