A new machine learning-powered algorithm has discovered eight new radio signals emanating from five stars 30 to 90 lightyears away. While we believe the signals are not from extraterrestrials, we cannot be certain.
Perhaps the aliens are communicating through “techno signatures,” which humans are only now discovering.
The algorithm’s creators, from the University of Toronto, claim to have “simplified the search for extraterrestrial life by employing a novel algorithm to organize data from their telescopes.” The algorithm filters out human-made interference using machine learning, allowing scientists to focus on real signals from deep space and then detect patterns in that information that could indicate they came from technologically manufactured signals.
“We need to distinguish exciting radio signals from space from uninteresting radio signals from Earth,” says Peter Ma, a Toronto undergraduate student and lead author of the new work, which appears in Nature Astronomy.
Of course, even if the transmissions appear to be extraterrestrial communications, experts say they aren’t confident aliens are behind them. However, they are expecting to detect the same signals again to raise the likelihood that these have additional meaning.
The researchers aren’t given up and plan to apply the new algorithm to a greater number of radio telescopes.
“With the help of artificial intelligence,” Cherry Ng, research associate at Toronto’s Dunlap Institute for Astronomy and Astrophysics, says in the news release, “I’m optimistic that we’ll be able to better quantify the likelihood of the presence of extraterrestrial signals from other civilizations.”