Alien-hunting AI discovers dozens of mysterious fast radio bursts
Deborah Byrd in HUMAN WORLD, SPACE | September 12, 2018
[video=youtube;BU1Df0oS1sg]https://www.youtube.com/watch?time_continue=3&v=BU1Df0oS1sg[/video]
Deborah Byrd in HUMAN WORLD, SPACE | September 12, 2018
[video=youtube;BU1Df0oS1sg]https://www.youtube.com/watch?time_continue=3&v=BU1Df0oS1sg[/video]
Researchers with Breakthrough Listen said on September 10, 2018, that they’ve now used artificial intelligence, AI, to discover 72 new fast radio bursts from a mysterious source some 3 billion light-years from Earth. The source is a known “repeater” of fast radio bursts – the only such object known to date – called FRB 121102.
Fast radio bursts, aka FRBs, are bright pulses of radio emission, lasting just milliseconds. They’re thought to originate in distant galaxies.
Most FRBs are one-offs, but this source – FRB 121102 – emits repeated bursts. FRB 121102 emitted 21 bursts previously detected during Breakthrough Listen observations made in 2017 with the Green Bank Telescope in West Virginia. Now – using AI – astronomers have found 72 more.
What are these mysterious bursts? Previous studies showed the bursts from 121102 emanating from a galaxy 3 billion light-years from Earth. Beyond that, we don’t know much about the source. Theories range from highly magnetized 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 civilization.
The 72 newly discovered FRBs are not about new observations. They’re about new and more powerful data-analysis techniques, made possible by artificial intelligence.
The statement from Breakthrough Initiatives continued:
In search of a deeper understanding of this intriguing object, the Listen science team at the University of California, Berkeley SETI Research Center observed FRB 121102 for five hours on August 26, 2017, using the Breakthrough Listen digital instrumentation at the [Green Bank Observatory]. Combing through 400 TB [terabytes] of data, they reported … a total of 21 bursts.
All were seen within one hour, suggesting that the source alternates between periods of quiescence and frenzied activity.
Now, UC Berkeley Ph.D. student Gerry Zhang and collaborators have developed a new, powerful machine learning algorithm, and reanalyzed the 2017 GBT dataset, finding an additional 72 bursts that were not detected originally.
Zhang’s team used some of the same techniques that internet technology companies use to optimize search results and classify images. They trained an algorithm known as a convolutional neural network to recognize bursts found with the classical search method … and then set it loose on the 400 TB dataset to find bursts that the classical approach missed.
Gerry 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 endeavors in applying machine learning to radio astronomy.