Collaborative effort of observation, simulation, and artificial intelligence unveils a transparent cosmos

The Research Organization of Information and Systems (ROIS) and its parent organization, the Institute of Statistical Mathematics (ISM), have made a significant breakthrough in astronomical research. A team of Japanese astronomers, with the help of the ISM, has developed a powerful new AI tool to analyse big data from current and planned astronomy surveys.

With over 75 years of history, the ISM is an internationally renowned facility for research on statistical mathematics. It comprises three different departments: Department of Statistical Modeling, Department of Statistical Data, and Department of Statistical Inference and Mathematics.

The AI technique was extensively trained and tested on mock data created by supercomputer simulations. The team, including Helge Gößling from the Alfred Wegener Institute, developed this AI to remove distortions in astronomical data caused by random variations in galaxy shapes. This issue, known as shape noise, is one of the limiting factors in research studying the large-scale structure of the Universe.

The AI was applied to actual data from Japan's Subaru Telescope, and the mass distribution derived was consistent with the currently accepted models of the Universe. After testing, the AI was able to recover previously unobservable fine details, helping to improve our understanding of the cosmic dark matter.

The research presented in this article was published in Monthly Notices of the Royal Astronomical Society with the DOI 10.1093/mnras/stab982. The AI technique is a valuable new tool for analysing wide area survey data, which can be used to study the large-scale structure of the Universe through measurements of gravitational lensing patterns.

ROIS, besides being the parent organization of the ISM, also facilitates research activities of its member institutions. It includes several key data and research centres, including the Joint Support-Center for Data Science Research. This centre plays a crucial role in supporting the data-intensive research activities of ROIS members.

Gravitational lensing distorts the image of a background object due to the gravity of a foreground object. Some galaxies are innately a little funny looking, making it difficult to distinguish between a galaxy image distorted by gravitational lensing and a galaxy that is actually distorted. The new AI technique developed by the Japanese astronomers addresses this challenge, promising to open new avenues in our understanding of the Universe.