In April 2019, a team of scientists released the famous first image of the supermassive black hole at the center of our galaxy. The black hole appears in the image with its thermal properties and shadow edges near which light bends as it travels through the universe. The hole appears as a glowing orange, yellow, and black disk, and appears as a dark dot on a dark ring.
But now, thanks to the development of artificial intelligence technology that we have witnessed, in April 2023, using machine learning algorithms, a team of scientists will be able to improve the image and make it twice as clear as it appears. Now there are black holes in the universe. . (M87) The center of the galaxy is deeper and larger than in the first image. This improvement, which was first observed in 2019, has important implications for measuring black hole masses.
But artificial intelligence is nothing new in astronomy, and astronomers have been using it for decades. 1990; Astronomers at the University of Arizona are pioneers in using a type of artificial intelligence called an artificial neural network (ANN) to study the shapes of galaxies.
Since then, artificial intelligence has permeated all areas of astronomy. Using powerful and advanced technology, artificial intelligence algorithms are beginning to help astronomers analyze huge data sets and discover new facts about the universe.
1. Better telescopes mean more data.
Astronomy was relatively simple, the only tools at that time were the naked eye or a simple telescope, and you could only see a few thousand stars and a few planets. But a hundred years ago, Edwin Hubble showed with a newly developed telescope that the universe is not only filled with stars and clouds of gas, but also with countless galaxies.
As telescopes improve, so does the number of celestial bodies that humans can see and the amount of data that astronomers need to collect and analyze.
For example: Chile's soon-to-be-completed Vera C. Rubin Observatory will capture images of an area so vast it would require 1,500 HDTV screens to display in its entirety over 10 years. The Library of Congress aims to produce 0.5 exabytes of data, which is 50,000 times more than the information contained in all the books.
Therefore, AI algorithms are the only way for astronomers to collect and analyze data available today, as AI is capable of quickly analyzing large amounts of data; It is therefore a very useful tool in decision making in various sectors.
2- Using artificial intelligence to identify patterns in galaxies:
Most tasks in exploration astronomy can be thought of as looking for a needle in a haystack. About 99% of the pixels in an astronomical image contain background radiation, light from other sources, or the darkness of space, and only 1% show the hidden shapes of galaxies.
Therefore, AI algorithms, especially neural networks, which use many interconnected nodes and are able to recognize patterns, are well suited for pattern recognition in galaxies. Astronomers began using neural networks to classify galaxies in the early 2000s, and the algorithms are now very efficient. It can classify galaxies with an accuracy of up to 98%.
This story is repeated in other areas of astronomy, as astronomers involved in the SETI project use radio telescopes to search for signals from distant civilizations.
Radio astronomers used to scan maps with the naked eye for unexplained anomalies, but recently, researchers used 150,000 computers to search for radio signals and artificial intelligence to analyze that data faster and more thoroughly. This allowed the SETI project to cover a larger area while greatly reducing the number of false signals.
Another example is the search for exoplanets. Astronomers have discovered most of the 5,347 known exoplanets by measuring the decrease in the amount of light emitted by a star as a planet passes in front of it. AI tools can now identify signs of exoplanets with up to 96% accuracy.
3. Artificial intelligence can help prove theoretical knowledge.
Artificial intelligence has proven to be an excellent solution for recognizing known objects. For example: galaxies and exoplanets, but it is also very powerful at finding things or phenomena that have evolved in theory but have not yet been discovered in the real world, as research teams use this method to discover new exoplanets and determine their resulting composition and growth. Intergalactic stellar fingerprints predict new types of gravitational waves. A 2015 demonstration of gravitational waves proved Einstein's theory, 100 years after it was proposed.
Finally, radio astronomers also use AI algorithms to search for signals that don't match known phenomena. A team from South Africa recently discovered a unique object that could be leftover from the merger of two supermassive black holes. If this turns out to be the case, the data will allow for further testing. Albert Einstein described the concept of spacetime in his general theory of relativity.
4. AI can help predict and close gaps:
As in many areas of modern life, generative AI and large language models (such as ChatGPT) have helped astronomers since the team that created the first image of a black hole in 2019 used "generative AI" to create their new image.
To do this, the team of scientists trained an AI model on how to detect black holes by simulating different types of black holes, and then the team used the AI models they created to fill in the gaps in the large amount of data. The data was collected by the M87 Black Hole radio telescope.
Using this data, the team was able to create a new image that is twice as sharp as the original, just as predicted by general relativity.
Astronomers are also turning to AI to manage search complexity. A team from the Harvard-Smithsonian Center for Astrophysics has created a language model called ASTROPRT to read and organize 15 million scientific papers on astronomy.
Another NASA team is also proposing to use artificial intelligence to prioritize astronomy projects, a process that astronomers engage in every decade.
finally:
Advances in artificial intelligence have made it an indispensable tool for astronomers, and as telescopes improve, data sets grow, and artificial intelligence develops, the technology is likely to play a pivotal role in the field of astronomy. Future cosmic discoveries.