Humans Can Still Find Galaxies That Machine Learning Algorithms Miss
The era of big data is upon us, and there are hardly any fields of scientific research that are not affected. Take astronomy for example. Thanks to the latest instruments, software and data sharing, observatories worldwide collect hundreds of Terabytes in a single day and between 100 to 200 Petabytes per year. Once the next generation of telescopes comes into operation, astronomy is likely to enter the “exabyte era”, where 1018 bytes (one quintillion) of data are acquired annually. To keep up with this volume, astronomers are turning to machine learning and AI to handle the task of analysis.
Although AI is playing a growing role in data analysis, there are some cases where civilian astronomers are more capable. While examining data collected by the Dark Energy Survey (DES), amateur astronomer Giuseppe Donatiello discovered three faint galaxies that a machine learning algorithm appeared to have missed. These galaxies, all satellites of the Sculptor Galaxy (NGC 253), are now named Donatello II, III and IV in his honor. In this day of data-driven research, it’s good to know that sometimes there’s no substitute for human eyeballs and intellect.
Right in the center of this image lies the newly discovered dwarf galaxy known as Donatiello II, one of three newly discovered galaxies. Credit: ESA/Hubble/NASA/B. Mutlu-Pakdil; Credit: G. Donatiello
The presence of these satellites around the Sculptor Galaxy (NGC 253), located 11.4 million light-years from Earth, has been confirmed by a team of astronomers using the Hubble Space Telescope. The team was led by Burçin Mutlu-Pakdil, an assistant professor of astrophysics at Dartmouth College (after whom Burçin’s Galaxy is named). The image below was part of a series of long exposure images of faint galaxies, showing Donatiello II in the center. The image has since become a Photo of the Week on the European Space Agency (ESA) website.
Remove all ads on Universe today
Join our Patreon for as little as $3!
Get the ad-free experience for life
Reliance on AI has increased significantly in recent years in direct response to the exponential increase in data obtained by astronomical observatories. In recent months, machine learning algorithms have been developed to search for exoplanets, fast radio bursts (FRBs), possible techno signatures and the mapping of the epoch known as Cosmic Dawn. But when it came to the DES, an international collaboration dedicated to mapping the cosmos to measure the nature and influence of Dark Energy, the algorithm they used was unable to detect these satellite galaxies.
This is not so surprising, since even the best algorithms have their limitations. To develop machine learning techniques, astronomers will train their algorithms using images and data from specific phenomena. Because some galaxies are so faint, AIs struggle to distinguish between them and individual stars and background noise. When this happens, identification must be done using the old-fashioned method of trained eyeballs combing through stacks of images and raw data.
Take that, Skynet!
Further reading: NASA, ESA