Artificial intelligence is unlocked the secrets of black holes

There may not However they are telescopes able to unlock all secrets of supermassive black holes, but ai is now on the case. Recently, an abnormia’s international team formed a nural network with millions of a similar black sketch to allow an unfair successful element in real life.

Of the various methods to investigate a black hole, horizon telescope event is the most famous. The EHT is not a single device but rather than a number of radio odesos in the world who work together as a single single telescope. Thank the eht, it was possible to get imagine of the supermassive black holes m87 and sagittarius a *. These have no images in the traditional sense but instead of radio waves arrive from the black holes.

To create these prints, supercomputors in different parts of the world transformed radio signals captured by the EHT. But in the process, they have picking up a lot in quantities that gather, it was difficult to interpret. The new neural network, trained in Wiscresin search Institute, purpose of tapping to that data to improve the resolution and make nine speech of EHT.

According to a picture release from the Institute, analyze artificial agirate of the update information and performed new sagittarius a *, how old the milky manner. An alternate image of the Burial Nero structure was generated, with this repaire, some new features of the black hole.

“Researchers are suspect the black hole at the milk speed center is pushed in nearly superior speed”, wrote the researchers in AA Picture of picture. I am The new image also indicates the black rotist assist on the ground and by their tracks the end and captures of the disks of the material.

The astronomers have been previously estimated that sagittarius a * rotate in a moderate speed. Knowing their current rotating speed, since it allows us to indifraising to the hacking to the hacking and providing the Romans.

“That differ you forgot theory of theory,” Leaven Cerneer Michael Jansen, of Radboud University NiJemy, in the Netherlands, they are in the Press free. “However, I see the IPLI and machine approach and a first step. After a first. Next, extend the associated patterns and simulations.”

This story has originally Wired in Spanish and has been translated by Spanish.

Source link