A University of Toronto student is among an international team of researchers using deep learning in the search for alien civilizations.
Artificial intelligence is now part of the search for extraterrestrial life.
Researchers have developed an artificial intelligence system that outperforms traditional methods in the search for extraterrestrial signals. And early results are intriguing enough to bring scientists back to their radio telescopes for a second look.
The research, published last week in Nature Astronomy, highlights the crucial role that artificial intelligence techniques will play in the ongoing search for extraterrestrial civilization.
The team behind the research has trained artificial intelligence to recognize signals that natural astrophysical processes cannot produce. They then fed it a massive data set consisting of over 150 terabytes of data collected by the Green Bank Telescope, one of the world’s largest radio telescopes located in West Virginia.
The system was created by Peter Ma, a graduate student at the University of Toronto and the lead author of the study, co-written by a constellation of experts affiliated with the University of Toronto, the University of California at Berkeley and Breakthrough Listen, an international facility launched in 2015 to search for signs of extraterrestrial civilizations.
Ma, who taught himself to code, first became interested in computer science in high school. He started working on a project that aims to use open-source data and solve problems with large amounts of data and unanswered questions, especially in the field of machine learning.
The second step feeds these features to a random forest classifier, which decides whether a signal is worthy of attention or just a disturbance.
The artificial intelligence system is particularly adept at identifying narrowband signals with non-zero outlier frequency. These signals are much more targeted and specific than natural phenomena and suggest they may be coming from a distant source.
Despite efforts dating back to the 1960s, only a small fraction of stars in the Milky Way have been observed. However, as technology advances, astronomers can now conduct more parallel observations and increase their scientific output.
Even the data collected, such as the Green Bank data, has not yet been fully explored.
And as next-generation radio telescopes, including MeerKAT, the Very Large Array (VLA), the Square Kilometre Array and the Next Generation VLA (ngVLA), collect vast amounts of data in the search for extraterrestrial intelligence, the application of AI will become increasingly important to overcome the challenges posed by the sheer volume of data.
