Skip to content

Unified effort of observation, simulation, and artificial intelligence reveal a detailed cosmic landscape

Artificial intelligence method innovated by Japanese researchers for minimizing noise in astronomical data caused by galaxy shape irregularities. After rigorous training and tests on data generated by supercomputer simulations, the new tool was employed on real data from Japan's Subaru...

Collaborative effort of observation, simulation, and artificial intelligence sheds light on the...
Collaborative effort of observation, simulation, and artificial intelligence sheds light on the cosmos, providing a lucid understanding.

Unified effort of observation, simulation, and artificial intelligence reveal a detailed cosmic landscape

The Institute of Statistical Mathematics (ISM), a renowned facility for statistical mathematics research, has played a pivotal role in a groundbreaking discovery by a Japanese research group. This group, led by astronomers from the National Astronomical Observatory of Japan, has developed an AI technique to remove noise in astronomical data, shedding new light on the mysterious cosmic dark matter.

ISM, part of Japan's Research Organization of Information and Systems (ROIS), comprises three different departments: the Department of Statistical Modeling, the Department of Statistical Data, and the Department of Statistical Inference and Mathematics. These departments have provided the theoretical foundation for the AI development.

The AI was designed to recover previously unobservable fine details, helping to improve our understanding of the cosmic dark matter. The team generated 25,000 mock galaxy catalogs based on real data from the Subaru Telescope and trained the AI to statistically recover the lensing dark matter from the mock data.

To compensate for shape noise, the team utilised ATERUI II, the world's most powerful supercomputer dedicated to astronomy. The new tool has been applied to actual data from Japan's Subaru Telescope and found to be consistent with the currently accepted models of the Universe.

The research presented in this article was published in the June issue of "Monthly Notices of the Royal Astronomical Society". The DOI for this article is 10.1093/mnras/stab982.

ROIS, a parent organization of four national institutes and the Joint Support-Center for Data Science Research, facilitates the research activities of its member institutions. This includes inter-university research institutes.

In gravitational lensing, the gravity of a foreground object can distort the image of a background object. Some examples of gravitational lensing are obvious, such as the "Eye of Horus." The large-scale structure, consisting mostly of mysterious "dark" matter, can distort the shapes of distant galaxies. Averaging over many galaxies in an area is required to create a map of foreground dark matter distributions.

The Institute of Statistical Mathematics (ISM), with more than 75 years of history, continues to contribute significantly to the advancement of astronomical research. This latest discovery is a testament to the institution's enduring commitment to pushing the boundaries of scientific knowledge.

Read also: