Published: Thu, September 13, 2018
Sci-tech | By Patricia Wade

Artificial Intelligence Detects 72 Mysterious Radio Bursts from Space

Artificial Intelligence Detects 72 Mysterious Radio Bursts from Space

In August of 2017, the Listen science team at the University of California, Berkeley SETI Research Center observed FRB 121102 for five hours, using digital instrumentation at the GBT. These bursts are extremely bright flashes of radio light that travel billions of light years from beyond our galaxy to reach Earth and appear momentarily on the sky. 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.

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, unusual and energetic pulses thought to emanate from far-off galaxies. While most FRBs are detected only during a single outburst, the fast radio bursts of 121102 repeat themselves, making it the only source that emits repeated bursts, including 21 detected bursts from last year's Green Bank Telescope (GBT) in West Virginia by Breakthrough Listen.

Fast radio bursts have puzzled astronomers since they were first detected around a decade ago. All had been seen within one hour, suggesting that the provide alternates between sessions of quiescence and frenzied process, acknowledged Berkeley SETI postdoctoral researcher Vishal Gajjar.

Now, UC Berkeley PhD 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. Thus, the discovery brought the total number of detected bursts from FRB 121102 to 300 since 2012. The technique the team used resembles algorithms used to optimize search results on search engines and to classify images.

"This work is only the beginning of using these powerful methods to find radio transients", said Zhang. Sure enough, the machine learning model picked out 72 more FRBs in the same period. They trained an algorithm known as a convolutional neural network to identify the earlier found FRB's, before setting it loose on the 2017 dataset. "We hope our success may inspire other serious endeavors in applying machine learning to radio astronomy".

FDA proposed crackdown on flavored e-cigs puts Juul on notice
And some of the retailers that received warning letters are still advertising and selling these products, he said. The FDA's delay on that requirement has allowed the industry to flourish with little oversight.

The new algorithm was very helpful in determining that source FRB121102 does not send out bursts at regular intervals (or at least not intervals longer than about 10 ms).

A paper on the research was recently accepted for publication in The Astrophysical Journal.

Step forward Listen - the initiative to get indicators of sparkling existence in the universe - announced today that a gape of thousands and thousands of stars located in the airplane of our galaxy, the use of the CSIRO Parkes Radio Telescope ("Parkes") ...

"This work is animated not impartial since it helps us perceive the dynamic habits of swiftly radio bursts in extra detail, however moreover thanks to the promise it reveals for the use of machine studying to detect signals missed by classical algorithms", acknowledged Andrew Siemion, director of the Berkeley SETI Examine Middle and fundamental investigator for Step forward Listen, the initiative to get indicators of sparkling existence in the universe.

Like this: