Scientists at the Hawaiian Astronomical Institute (IfA) have used artificial intelligence to create the world’s largest 3D catalog of known stars, galaxies and quasars. To do this, astronomers used optical data from three-quarters of the sky, collected by the Pan-STARRS observatory on the island of Maui.
To train the AI, spectroscopic data from 4 million light sources were used. This gave the neural network the ability to determine the type of source and the distance to it, while simultaneously correcting for light scattering by cosmic dust. Thanks to this, the classification accuracy of objects has reached 98.1% for galaxies, 97.8% for stars and 96.6% for quasars. The resulting astronomical catalog is twice the size of the previous record holder, which was created as part of the Sloan Digital Sky Survey (SDSS) and covers only one third of the sky.
Already the very first, preliminary version of the new catalog made it possible to make an amazing discovery – to discover the largest area of emptiness in the known universe. It is likely that it is he who is the cause of the so-called relict cold spot or super-void Eridanus. This huge and unusually cold region of space in the constellation Eridanus casts doubt on the standard cosmological model and has long been a stumbling block for astrophysicists.
The full 300 GB catalog can be downloaded from this site.