[Apr. 8, 2023: Staff Writer, The Brighter Side of News]
They discovered eight extraterrestrial signals that appear to have the characteristics of the technology. (CREDIT: SETI Institute)
Are we alone in the universe?
Scientists may have just brought us closer to the answer to that question. The team – led by researchers from the University of Toronto – streamlined the search for extraterrestrial life by using a new algorithm to organize data from their telescopes into categories, to distinguish between real signals and interference. This allowed them to quickly sort through information and find patterns, thanks to an artificial intelligence process called machine learning.
They discovered eight extraterrestrial signals that appear to have the characteristics of the technology. The study, published in the journal Nature Astronomydoes not claim to have found evidence of intelligent extraterrestrials, but researchers believe that using artificial intelligence is a promising way to search for extraterrestrial intelligence.
“I am impressed with the performance of this approach in the search for extraterrestrial intelligence,” said the study’s co-author. Cherry Ngan astronomer from the University of Toronto, said in a statement. “With the help of artificial intelligence, I am optimistic that we will be able to better quantify the likelihood of the presence of extraterrestrial signals from other civilizations.”
The search for extraterrestrial intelligence, or SETI, has been underway since the 1960s and focuses on finding evidence of technology-generated signals, known as technosignatures, from advanced extraterrestrial civilizations. Astronomers use powerful radio telescopes to scan thousands of stars and hundreds of galaxies in hopes of discovering these technosignatures. It is assumed that an advanced extraterrestrial civilization would possess the necessary sophistication to emit such signals.
Related stories
Despite being located in areas with minimal interference from technology, the search for extraterrestrial intelligence (SETI) still faces major challenges due to human disturbances. Peter Ma, an undergraduate student and researcher at the University of Toronto, explains that “in a lot of our observations, there’s a lot of interference.”
To differentiate extraterrestrial signals from human-generated interference, the team trained their machine learning tools through simulations of both types of signals. They tested a variety of algorithms, evaluated their accuracy and false positive rates, and finally settled on a powerful algorithm created by Ma.
The new technique uses a method called “semi-unsupervised learning”, which combines supervised and unsupervised learning. The algorithm was first trained to differentiate between man-made radio signals originating on Earth and signals originating elsewhere. Researchers analyzed 150 terabytes of data from the Green Bank Telescope in West Virginia, covering observations of 820 stars near Earth, and found eight previously overlooked signals from five stars located between 30 and 90 light-years from Earth.
Artist’s impression of the Green Bank Telescope connected to a machine learning network. (Credit: Breakthrough Listen/Danielle Futselaar)
Ma’s algorithm, called “semi-unsupervised learning”, is a combination of two subtypes of machine learning, supervised and unsupervised learning. It uses the strengths of both techniques to improve the accuracy of the algorithm. In this approach, supervised learning is used to guide and train the algorithm, while unsupervised learning is used to uncover hidden patterns in the data. This combination allows the algorithm to generalize the information it has learned and more easily detect new patterns in the data, leading to better results in finding extraterrestrial signals.
Ma’s innovative idea of applying semi-unsupervised learning to SETI began as a high school project. “I didn’t tell my team until after the newspaper was published that it all started as a high school project that wasn’t really liked by my teachers.”
Dr Ng says new ideas are very important in a field like SETI. “By digging into the data with each technique, we might be able to uncover some exciting signals.”
U of T student and researcher Peter Ma. (Polina Teif)
Scientists from the Breakthrough Listen SETI effort say these signals had two characteristics in common with signals that could be created by intelligent aliens: they are present when looking at the star and absent when looking away, and they change frequency over time in a way that causes them to appear far from the telescope. However, these features could arise by chance and further observations are needed to make any statements about extraterrestrial life.
“First, they are present when we look at the star and absent when we look away – as opposed to local interference, which is usually always present,” Steve Croft (opens in a new window), Breakthrough Listen project scientist at Green Bank Telescope, said in the release. “Second, the signals change frequency over time in a way that makes them appear far from the telescope.”
The Green Bank Telescope. (Credit: Chris Schodt/Breakthrough Listen)
The research team hopes to apply its algorithm to data from more powerful radio telescopes, such as MeerKAT in South Africa or the Next Generation Very Large Array project. They believe this new technique, combined with the next generation of telescopes, will allow them to search for millions of stars instead of just hundreds.
“With our new technique, combined with the next generation of telescopes, we hope that machine learning can take us from searching for hundreds of stars to searching for millions,” Ma said.
Although the first results do not lead to the discovery of extraterrestrial life, the use of machine learning in the search for extraterrestrial intelligence is very promising. The study authors are optimistic that artificial intelligence will help them better quantify the likelihood of the presence of extraterrestrial signals from other civilizations.
For more scientific news, see our New Innovations section on The bright side of the news.
Note: The documents provided above by The bright side of the news. Content may be edited for style and length.
Do you like these kind of wellness stories? Get the Brighter Side of News Newsletter.