Artificial intelligence could be the best tool to explore the universe

Space robots can help us better understand the cosmos.

For some astronomers, the potential for the development and evolution of artificial intelligence (AI) goes more beyond mere data classification.They believe that artificial intelligence could be used to create information, filling the blind points in our observations of the universe.

A group of astronomers from the Californian University of Berkley, who specialize in astrophysics of galaxies and black holes, have used artificial intelligence to improve the resolution of the blurred images of the telescope.To do this, they deployed a type of neuronal network that is optimal to generate variations of the data you study, as a well -trained counterfeit that can imitate the style of a famous painter.These networks, called generative confrontation networks, or GAN (for its acronym in English), have been used to create false faces based on celebrity images;False audio dialogue that mimics people's voices;and a wide variety of other types of data.They are one of the most developed variants of research in contemporary artificial intelligence and, for researchers it means obtaining information that did not exist before.

Windows 10 on the family computer, day 1 "I didn't know how to change the sound output, so I UNINSTALLED ALL THE AUDIO DRIVERS." -My sister

— James Thu Jul 30 03:46:22 +0000 2015

Imitates colors like a famous painter

The article published by this group of researchers from Berkley at the end of 2021, showed how generative confrontation networks could be used to improve the quality of space images.They lowered the image quality of a lot of galaxies images, adding noise and blur, then used a generative networks of specialized confrontation in telescope images to increase their resolution, comparing them with the originals.The results were surprisingly precise: good enough to convince the scientific community that studies the cosmos that artificial intelligence has a great potential to improve all types of data sets in the field of astronomy.The scientists involved in the project ensure that they have "many interesting results in process", but cannot reveal anything before they are published in the usual specialized scientific publications.

Even so, researchers are still very cautious with the project.After all, it seems that it goes against the fundamental principles of science: that you can only learn about the universe observing it directly."This is a dangerous tool precisely for this reason," said the group of scientists of the University, and should only be used when we have broad and precise databases, and we can verify the results.

Therefore, a generative confrontation network can be used to generate data on black holes and then release it in a part of the sky that has not been observed in much detail before.Then, if the data suggests that there is a black hole there, astronomers will confirm this first -hand finding, as with gravitational lenses.Scientists say that, as with all scientific tools, it is necessary.

Similar contents