One of astronomy’s controversial mysteries is now being investigated by artificial intelligence.
A team of researchers at Breakthrough Listen, a Search for Extraterrestrial Intelligence (SETI) project spearheaded by the University of California, Berkeley, have developed a machine learning algorithm to sift through cosmic data and identify fast radio bursts, strange and energetic pulses thought to emanate from far-off galaxies. No one knows for sure what causes these radio emissions but theories abound — from highly magnetized neutron stars battling black holes to signs of alien life.
In a recent study, the SETI researchers used a customized A.I. system to discover dozens of previously unidentified fast radio bursts from a source some 3 billion light-years away. The fast radio bursts were picked out from a data set that had already been analyzed by astronomers.
“Artificial Intelligence has seen very rapid development in recent years and its application in medicine, security, finance, and everyday object recognition have already reached certain level of maturity,” Gerry Zhang, a UC Berkeley Ph.D. student who helped develop the new machine-learning algorithm, told Digital Trends. “Astronomy represents a relatively unexplored area for A.I. The very large volume of data that astronomers collect present a challenging playground for state-of-the-art A.I. Radio astronomical data themselves presents challenges of large scale and high noise. Learning to apply A.I. to such new data is challenging and fruitful.”
To make the discovery, Zhang and his team used a convolutional neural network, a type of algorithm modeled off the human brain, which has been used to find craters on the moon and help detect earthquakes.
In their recent study, the researchers trained their algorithm on simulated signals, teaching it to recognize signs of fast radio bursts, and then “let the trained network loose on the data containing the real signals,” Zhang said. The result was the discovery of 72 signals that astronomers had previously missed.
The new findings help chip away at the current cosmic mystery of where the fast radio bursts come from. Is it aliens? Probably not. But it could be. Either way, it’s a mystery worth getting to the bottom of. What astronomers need now is more data and better systems to analyze the information.
“Fast radio bursts are one of the most recently discovered unknown signals in astronomy,” Zhang said. “With new instruments being designed for them coming online, [they are] well-poised to be one of the unknowns that will be solved within the next five to 10 years.”
A paper on the research was recently accepted for publication in The Astrophysical Journal.
- Machine learning? Neural networks? Here’s your guide to the many flavors of A.I.
- The fastest cars in the world
- By studying patient data, A.I. can limit toxicity in cancer treatment
- Don’t be fooled — this automated system sneakily manipulates video content
- Like a vice principal in the sky, this A.I. spots fights before they happen