Breakthrough Listen, the astronomical program responsible for searching for signs of intelligent life in the universe, has detected 72 new fast radio bursts (FRBs) that could have come from distant galaxies.
Using the latest machine learning techniques, the programme found the FRBs emanating from a "repeater" called FRB 121102.
Fast radio bursts, or FRBs, are bright pulses of radio emission, just milliseconds in duration, thought to originate from distant galaxies. Most FRBs have been witnessed during just a single outburst.
However, FRB 121102 is the only one to date known to emit repeated bursts. And the source of FRBs are still a mystery to astronomers.
Theories range from highly magnetised neutron stars, blasted by gas streams near to a supermassive black hole, to suggestions that the burst properties are consistent with signatures of technology developed by an advanced civilisation.
However, the discovery doesn't come from a new observation. The Listen science team at the University of California, Berkeley SETI Research Center, originally observed FRB 121102 on 26 August last year, using the Breakthrough Listen digital instrumentation.
Combing through 400 terabytes of data, they reported a total of 21 bursts. All of which were seen within one hour.
Now, University of California at Berkeley PhD student Gerry Zhang and collaborators have developed a new, powerful machine learning algorithm, and re-analysed the 2017 GBT dataset, finding an additional 72 bursts that were not detected originally.
"Not all discoveries come from new observations," said Pete Worden, executive director of the Breakthrough Initiatives. "In this case, it was smart, original thinking applied to an existing dataset. It has advanced our knowledge of one of the most tantalising mysteries in astronomy."
Zhang's team used some of the same techniques that internet technology companies use to optimise search results and classify images. They trained an algorithm known as a 'convolutional neural network' to recognise bursts found with the classical search method used by Gajjar and collaborators, and then set it loose on the 400 TB dataset to find bursts that the classical approach missed.
The results have helped put new constraints on the periodicity of the pulses from FRB 121102, suggesting that the pulses are not received with a regular pattern.
"This work is only the beginning of using these powerful methods to find radio transients," said Gerry Zhang. "We hope our success may inspire other serious endeavours in applying machine learning to radio astronomy."
Could be used for everything from search-and-rescue robots to wearable tech
Don't require the rare material being mined from the mountains of South America
IBM hopes that its new tool will avoid bias in artificial intelligence
Found by calculating the strength of the material deep inside the crust of neutron stars