The Search for Extraterrestrial Intelligence (SETI) Institute has spotted several radio burst coming from a distant galaxy as part of their Breakthrough Listen initiative. It is a project setup specifically to look for and find "signs of intelligent life in the universe".

In a new finding, a release put out by SETI has confirmed that researchers applied machine learning and AI have spotted 72 new fast radio burst (FRB) that seem to copy "repeater" FRB 121102. FRBs, explains SETI are pulses of radio emission. They last only a few milliseconds at a time and they are believed to originate from distant galaxies.

While FRBs for the most part have been recorded with just one outburst, FRB 121102 is the only one, known till now to repeat bursts. This includes one episode where 21 outbursts were recorded in 2017 using the Green Bank Telescope (GBT) in West Virginia as part of a Breakthrough Listen study.

Gerry Zhang of the UC Berkeley, lead author of the paper that published the findings this week. The report notes that the new findings did in fact come out of FRB 121102, but were not detected by researchers. This is where the AI and ML program comes to play. The bursts came from a dwarf galaxy, researchers find, that is about 3 billion light years away. However, as to what created the signals, scientists are yet to find out. Theories, are aplenty and one of them is that it could have come from an extraterrestrial scientifically advanced civilization and that it could be the signature of their technology.

alien, cosmic light flares, space,
Mysterious radio bursts are being linked to advanced alien technology by researchers.CfA / Harvard-Smithsonian Center for Astrophysics

The signal was first picked up in August 2017 by a Breakthrough Listen team working out of the UC Berkeley SETI Research Center. The recorded FRB 121102 lasted for five hours, using the GBT.

The five hour FRB session generated 400 TB of data and the team was able to report 21 bursts. All of them within a span of one hour, nots the report. This suggests that the source alternated between frenzied outbursts interspersed with short periods of calm.

A new machine learning algorithm developed by the team was put to work re-analyzing the data dump. It was in it that they uncovered an additional 72 bursts to the original 21. Known as a convolutional neural network, it was trained to recognize FRBs on the same principles put forwad by Gajjar and collaborators—the team that found and published the first find—researchers then set the AI loose on the entirety of the 400 TB to see if there was anything that human eyes might have missed out on.

"These results hint that there could be vast numbers of additional signals that our current algorithms are missing and clearly demonstrate the power of applying modern data analytics and AI tools to astronomical research," said SETI Institute President and CEO Bill Diamond.

Applying these techniques, he explained, in the search for evidence of extraterrestrial technologies, or technosignatures, is incredibly compelling, together with addressing the tantalizing phenomena of FRBs.

Results of this research will be published in the Astrophysical Journal it is also now available on the arXiv service to explore.

AI finds alien signals
New detections mark the first time that machine learning techniques have been used to directly detect a fast radio transientBreakthrough Listen / Danielle Futselaar