Japanese astronomers have developed a new artificial intelligence (AI) technique to remove noise in astronomical data due to random variations in galaxy shapes. After extensive training and testing on large fictitious data created by supercomputer simulations, they then applied this new tool to real data from the Japanese Subaru telescope and found that the mass distribution derived from using this method is consistent with currently accepted models of the Universe. This is a powerful new tool for analyzing big data from current and planned astronomical surveys.
Large-scale survey data can be used to study the large-scale structure of the Universe through measurements of gravitational lens models. In a gravitational lens, the gravity of a foreground object, such as a cluster of galaxies, can distort the image of a background object, such as a more distant galaxy. Some examples of gravitational lenses are obvious, such as the “Eye of Horus”. The large-scale structure, made mostly of mysterious “dark” matter, can also distort the shapes of distant galaxies, but the expected lens effect is subtle. An average over many galaxies in an area is needed to create a map of foreground dark matter distributions.
But this technique of looking at many images of galaxies comes up against a problem; some galaxies are just a little bit funny by nature. It is difficult to distinguish between an image of a galaxy distorted by a gravitational lens and a galaxy that is actually distorted. This is called shape noise and is one of the limiting factors in research into the large-scale structure of the Universe.
To compensate for shape noise, a team of Japanese astronomers first used ATERUI II, the world’s most powerful supercomputer dedicated to astronomy, to generate 25,000 catalogs of fake galaxies based on real data from the Subaru telescope. . They then added realistic noise to these well-known man-made data sets and trained an AI to statistically recover lens dark matter from the dummy data.
After the training, the AI was able to retrieve fine details previously unobservable, helping to improve our understanding of cosmic dark matter. Then using this AI on real data spanning 21 square degrees of the sky, the team found a prominent mass distribution consistent with the standard cosmological model.
“This research shows the benefits of combining different types of research: observations, simulations and analysis of AI data.” comments Masato Shirasaki, team leader, “In this age of big data, we need to move beyond traditional boundaries between specialties and use all available tools to understand data. If we can do that, it will open up new areas. in astronomy and other sciences. “
Source of the story:
Material provided by National institutes of natural sciences. Note: Content can be changed for style and length.