Artificial intelligence is infiltrating many scientific fields, including astronomy and the search for intelligent life in the universe known as SETI.
In a new study, scientists at Breakthrough Listen, a SETI project led by the University of California at Berkeley, USA, used machine learning algorithms to discover 72 new fast radio bursts from a mysterious source about 3 billion light years from Earth. …
Fast radio bursts are powerful pulses of radio waves only a few milliseconds long, which are believed to originate from distant galaxies. The sources of this radiation, however, are still unknown to scientists. The proposed explanations for these mysterious flares in the radio range vary from neutron stars with powerful magnetic fields, attacked by jets of nearby black holes, to versions that include the idea of technically advanced civilizations sending us their signals.
In the new work, a team led by UC Berkeley PhD student Gerry Zhang successfully developed a new, powerful machine learning algorithm and applied it to study FRB 121102, a unique event of multiple, repetitive radio bursts recorded in 2012 with a telescope Green Bank, located in West Virginia, USA. Using this new algorithm allowed the team to detect an additional 72 fast radio bursts that had not been previously detected. As a result, the total number of detected fast radio bursts of the FRB 121102 event is now 300 flares.
These results also helped Zhang's team to impose new restrictions on the frequency of fast radio bursts - the work showed that there is no periodicity in the arrival of pulses, at least for periods longer than 10 milliseconds.
The study is accepted for publication in the Astrophysical Journal.