Fast radio bursts (FRBs) are a growing area of interest for the world’s expert astronomers. FRBs manifest as short blasts of strong radio waves, emanating from deep space.
One of 2017’s most intriguing astronomical puzzles centred around the discovery of 15 FRBs from a galaxy 3bn light years away. FRBs were first detected in 2012, but the 2017 instance showed that the bursts emanated from the ‘repeater’ FRB 121102.
The same team that discovered the FRBs in 2017, Breakthrough Listen, has continued its work over the last few months. The project operates under the Search For Extraterrestrial Intelligence (SETI) Institute and is based out of the University of California, Berkeley. Researchers there have been able to harness machine learning (ML) to discover a whopping 72 FRBs from 3bn light years away.
Using ML to track fast radio bursts
Scientists used a new ML algorithm to re-examine data from the FRB 121102 repeater source. The original data had been acquired in 2017 by the Green Bank Telescope in the US state of West Virginia. The vast reams of information were originally analysed using traditional methods.
Gerry Zhang, a doctoral student at the university, trained an algorithm called a ‘convolutional neural network’ to flag FRBs in the 400TB of data. The total number of FRBs from the 121102 repeater is now 93.
Zhang said: “This work is only the beginning of using these powerful methods to find radio transients. We hope our success may inspire other serious endeavours in applying machine learning to radio astronomy.”
Andrew Siemion, director of the Berkeley SETI Research Center and principal investigator for Breakthrough Listen, said: “This work is exciting not just because it helps us understand the dynamic behaviour of fast radio bursts in more detail, but also because of the promise it shows for using machine learning to detect signals missed by classical algorithms.”
Pete Worden, executive director of Breakthrough Initiatives, which includes Breakthrough Listen, said: “Not all discoveries come from new observations. 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.”
According to the researchers, the source of these emissions is still not clear. One theory involves highly magnetised neutron stars hit by gas streams from a supermassive black hole nearby. Others suggest that the burst properties are consistent with signatures of technologies developed by advanced civilisations.
The post Astronomers identify 72 new fast radio bursts with a little help from AI appeared first on Silicon Republic.