AI Used in Search for Extraterrestrial Life:

Over at Vice, Sarah Wells tell us that “AI Has Found Potential Alien ‘Technosignatures’ Hidden in Radio Signals From Space.”

Peter Ma, first author on a paper about machine learning and the search for extraterrestrial life, writes:

Previously people have inserted ML components into various pipelines to help with the search. This work relies entirely on just the neural network without any traditional algorithms supporting it and produced results that traditional algorithms did not pick up.

OK, so AI has been applied to SETI’s radio data in the past. However, this new approach takes the search entirely out of human hands.
In 2016, Ma and his team crafted a neural network to classify the most critical features of SETI data, while attempting to weed out any Earth-based interference. The machine-learning algorithm analyzed 480 hours of data from 820 stars. It identified eight signals of interest that previous algorithms had not detected. Nothing yet about Chuck Berry.

Author