Astronomers Unveil Largest 3D Catalog of Galaxies Ever

The IfA researchers gathered publicly available spectroscopic measurements before feeding it to an artificial intelligence (AI) algorithm.

The AI system used a "feedforward neural network" to achieve an overall classification accuracy of 98.1% for galaxies, 97.8% for stars, and 96.6% for quasars. Galaxy distance estimates, meanwhile, are accurate to almost 3%.

"Utilizing a state-of-the-art optimization algorithm, we leveraged the spectroscopic training set of almost 4 million light sources to teach the neural network to predict source types and galaxy distances, while at the same time correcting for light extinction by dust in the Milky Way," Lead study author Robert Beck, a former cosmology postdoctoral fellow at IfA, explained in a press release.

The largest catalog of galaxies

The record for the largest map of the universe was previously held by the Sloan Digital Sky Survey (SDSS), which covers one-third of the sky. The PS1 3π survey doubles the area surveyed in the SDSS.

As IfA astronomer and co-author on the study, István Szapudi, points out that "a preliminary version of [the PS1 3π catalog], covering a much smaller area, facilitated the discovery of the largest void in the universe, the possible cause of the Cold Spot. The new, more accurate, and larger photometric redshift catalog will be the starting point for many future discoveries."

Pan-STARRS Director and IfA Associate Astronomer, Ken Chambers, says that "this beautiful map of the universe provides one example of how the power of the Pan-STARRS big data set can be multiplied with artificial intelligence techniques and complementary observations."

The neatly-cataloged dataset will help astronomers and AI to find even more information about near-Earth objects and may well lead to the uncovering of many of the mysteries of our universe.