Machine learning is unlocking the cosmos.
Scientists from the National Astronomy Meeting announced that humans worked together with machine intelligence to find 40,000 ring galaxies throughout the universe, a press statement reveals.
Dr. Mike Walmsley of the University of Manchester and the Galaxy Zoo collaboration developed the so-called “cyborg” method to measure the shapes of millions of different galaxies.
New Zoobot AI helps identify thousands of ring galaxies
The Galaxy Zoo website, a citizen science initiative, collects data on stars whose orbits have been disrupted by galactic collisions and bursts of energy from supermassive black holes.
However, categorizing all of this information based on the specific cosmic event that triggered each massive disruption requires millions of measurements. It totals to such an overwhelming amount of data that it could take human volunteers several lifetimes to sort through. That’s where machine intelligence comes in.
Dr. Walmsley used a decade of Galaxy Zoo volunteer measurements to create a new AI algorithm, named “Zoobot,” that could work as an assistant on the project. It is able to accurately predict what human volunteers would say about specific data points, as well as understand where it might be mistaken.
Zoobot uses machine learning, meaning it is trained over and over again on data set until it can carry out the required function at incredible speeds. According to Dr. Walmsley, “with Zoobot, humans and machines are collaborating to push the science of astronomy forward. We’re helping other astronomers solve questions we never thought to ask.”
Could machine intelligence help us find alien intelligence?
Thanks to the Zoobot AI, Dr. Walmsley was able to discover 40,000 rare ring-shaped galaxies — six times more than all previously known ring galaxies. The rare galaxy type takes billions of years to form and they are destroyed during relatively common galaxy-galaxy collisions. The massive new dataset, therefore, will help scientists investigate how these isolated galaxies evolve.
The work on Zoobot, as well as other machine learning astronomical solutions, will help scientists to sort and investigate massive amounts of data in the future. Back in 2018, for example, researchers from Plymouth University identified how artificial intelligence could be leveraged to help in the search for extraterrestrial intelligence — their algorithm was able to quickly identify planets with habitable conditions.
“Galaxy Zoo turns 15 years old this week, and we are still innovating,” said Galaxy Zoo Deputy Principal Investigator Dr. Brooke Simmons, of the University of Lancaster. “The work Dr. Walmsley is leading will make it possible for a new generation of discoveries to be made from upcoming large galaxy surveys.”
In fact, machine learning is already used extensively in the field of astronomy. Just this week, for example, researchers from Australia’s Curtin University announced they pinpointed the exact location, on Mars, where the Black Beauty meteorite was born. Researchers from the University of California, Santa Cruz, are also currently training a machine learning AI called Morpheus on that first stunning James Webb image revealed earlier this week.