Are we the only ones on the planet?
Scientists might have moved closer to solving this issue. The team of researchers at Toronto's University of Toronto - has made it easier to hunt for extraterrestrial life applying a new method to arrange the data collected by the telescope into groups so that they can distinguish between genuine signals and interference. This allows them to swiftly analyze the data and identify patterns using the process of artificial intelligence, also known by the term machine learning.
The researchers discovered eight signals from the universe that appear to bear the characteristics of technological. The study, which is published within the journal Nature Astronomy, doesn't claim to have discovered proof of intelligence from aliens however, the researchers believe that using artificial intelligence could be an exciting way to look for intelligence from beyond.
"I am impressed by how well this approach has performed on the search for extraterrestrial intelligence," co-author of the study Cherry Ng, an astronomer at the University of Toronto, said in an announcement. "With the help of artificial intelligence, I'm optimistic that we'll 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 ongoing since the 1960s and is focused on finding evidence of technologically-generated signals, known as techno signatures, from advanced extraterrestrial civilizations. Astronomers have used large radio telescopes in order to look through thousands of galaxies and hundreds of stars to try and discover the techno signatures. It is thought that an advanced extraterrestrial civilization has the capability to produce such signals.
Even though they're located in regions that are not impacted by technology the hunt for alien intelligence (SETI) has to face significant issues due to human disturbance. Peter Ma, an undergraduate student and researcher at the University of Toronto, explains that "in many of our observations, there is a lot of interference."
To distinguish the signals of extraterrestrials from interference generated by humans The team taught their machine-learning programs by simulated both kinds of signals. They evaluated a variety of algorithms, assessed their accuracy and false-positive rate and finally settled on the most effective algorithm designed by Ma.
The latest technique employs the method known as "semi-unsupervised learning," which blends unsupervised and supervised learning. The algorithm was taught to distinguish between radio signals coming from Earth and signals from other. Researchers analyzed 150 terabytes of data gathered from the Green Bank Telescope in West Virginia that covered observations of 820 stars close to Earth and identified eight previously unnoticed radio signals from five star systems that lie in between 30-90 light years away from Earth.
Ma's algorithm, often referred to in the context of "semi-unsupervised learning," is the result of combining two subtypes of machine learning: unsupervised and supervised learning. It makes use of advantages of each methods to increase the efficiency in the process. This method employs the use of supervised learning to help guide and train the algorithm and unsupervised learning is employed to find hidden patterns in the data. This lets the algorithm generalize the knowledge it has acquired and more quickly recognize new patterns within the data, resulting in more effective results when searching for signals from the outer space.
Ma's original idea of applying semi-unsupervised teaching to SETI was originally an idea for a high school assignment. "I only told my team after the paper's publication that this all started as a high-school project that wasn't really appreciated by my teachers."
Dr. Ng, says new ideas are essential in fields like SETI. "By poking the data with every technique, we might be able to discover exciting signals."
Researchers of Breakthrough Listen SETI say Breakthrough Listen SETI effort say that these signals had two things similar to signals that could be made by intelligent aliens. They appear when you look at the star but not visible when looking away. Furthermore, they shift in frequency with time in a manner that causes them to appear distant distance from the telescope. But, these characteristics could be a result of chance, and further studies are required for any assertions 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 generally always present," Steve Croft(opens in a new tab) the project scientist of Breakthrough Listen on the Green Bank Telescope, said in the statement. "Second, the signals change in frequency over time in a way that makes them appear far from the telescope."
The research team is hoping to apply their method to data from higher-powered radio telescopes, like MeerKAT located in South Africa or the planned Next Generation Very Large Array. They believe that this innovative method, in conjunction with the next generation of telescopes will allow them to study hundreds of thousands of stars instead of hundreds.
"With our new technique, combined with the next generation of telescopes, we hope that machine learning can take us from searching hundreds of stars to searching millions," Ma declared.
Even though the initial results aren't concluding the existence extraterrestrial life using machine learning to search for extraterrestrial intelligence has great potential. The researchers behind this study are hopeful that artificial intelligence can help to determine the probability of finding extraterrestrial signals that originate from other civilizations.
Tags
announcement
artificial intelligence
civilization
distant stars
effective algorithm
extraterrestrial life
method of data
News
observation
presence of signals
scientists discover
technological