Alien-Hunting AI Catches 72 Mysterious Radio Bursts From Outer Space
Artificial intelligence helps with the search for intelligence life by detecting an astonishing number of mysterious fast radio bursts from 3 billion light-years away.
AI hunting for aliens — it sounds straight out of a science fiction movie. For astronomers, artificial intelligence paves the way for greater, more efficient methods in their operations, including the search for life outside of Earth.
Dozens Of Radio Bursts From A Mystery Source
According to SETI Institute, their researchers have used machine learning techniques to capture 72 new FRBs from the repeating FRB 121102 in a dwarf galaxy 3 billion light-years from Earth. FRB 121102 is the only known FRB to emit repeated pulses, as most just give off a single outburst.
Fast radio bursts or FRBs are bright flashes of radio emission that are thought to have come from distant galaxies. The nature of their source is still unknown. There are multiple theories on FRB origins, with some suggesting that they could be traces of technology used by intelligent life somewhere in the universe.
The search for FRBs is part of the institute's Breakthrough Listen project, which is an initiative led by the University of California, Berkeley, dedicated to look for signs of extraterrestrial life.
Machine Learning Boosts The Search For Intelligent Life
The most significant finding of the research, which has been accepted for publication in the Astrophysical Journal, is the potential of machine learning techniques in Breakthrough Listen's work.
"This work is exciting not just because it helps us understand the dynamic behavior 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," Andrew Siemion, director of the Berkeley SETI Research Center and principal investigator for Breakthrough Listen, explains in a report from UC Berkeley.
In August 2017, the team observed FRB 121102 for a span of five hours with digital instrumentation. In the 400 TB of data from that period, the researchers found a total of 21 bursts. All were detected within an hour, suggesting the source alternates between quiet and frenzied activity.
By developing a new machine learning algorithm, lead author and UC Berkeley postdoctoral student Gerry Zhang and other researchers were able to comb over the same data from 2017 and find an additional 72 bursts that the classical approach missed. Specifically, the team trained an algorithm to identify the bursts found by the 2017 team and then used it on the dataset to find other similar bursts that weren't found back then.
"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," Bill Diamond, SETI Institute President and CEO, says. "Applying these techniques in the search for evidence of extraterrestrial technologies, or technosignatures, is incredibly compelling, together with addressing the tantalizing phenomena of FRBs."