Machine learning could boost the search for extraterrestrial intelligence

Posted on

The field of SETI, or the search for extraterrestrial intelligence, is accelerating to new heights thanks to developments in machine learning.

In an article that appeared last week in Nature Astronomya team of researchers led by Peter Ma of the University of Toronto shared a machine learning method for sifting through data from the Breakthrough Listen project to identify signals that could be potential technosignatures – that is, indications of technological complexity that suggest intelligent alien civilization.

“I think it’s a really important question to ask as humanity: Is there anyone else?” Mom told The edge.

“I think it’s a really important question to ask as humanity: Is there anyone else?”

As humanity becomes increasingly adept at observing the universe and learning about its history, the question of whether we are alone has never been sharper. If there is life beyond Earth, how can we find it? Why haven’t we been contacted yet? And what would it take to connect with an alien civilization?

Ma’s research focuses on a particular part of the electromagnetic spectrum called narrowband radio. While all kinds of objects in the universe emit radiation with a wide range of frequencies, radio frequencies are particularly efficient for transmitting signals. And when we as humans communicate through radio waves, we use a narrow band because it’s more efficient.

SETI researchers assume that alien civilizations, if out there, would do the same. “From a technology perspective, it makes sense that any intelligent civilization, which also tries to broadcast through electromagnetic radiation, such as radio, would broadcast on narrow bands,” Ma explained.

Why haven’t we been contacted yet? And what would it take to connect with an alien civilization?

There is a certain part of the radio band, around the 1420 MHz range, that SETI researchers are interested in. Known as the hydrogen line, this is important to astronomers because it is the frequency at which neutral hydrogen emits radiation, so it is key to studying all sorts of astronomical goals.

Researchers believe that any extraterrestrial civilization interested in the stars would likely also be watching this tape, making it what has been called a “galactic watering hole.” If a civilization were to try to communicate throughout the cosmos, that’s the best idea we have for the frequency they would use to do so.

This approach has been the basis for much SETI research in previous years: digging through data to look for signals in this band, often using an algorithm called turboSETI. That algorithm searches plots of time by frequency and looks for straight lines, which indicate the presence of a signal. It’s an efficient way to sift through masses of data, but it also has problems – not least of which is filtering out false positives caused by interference from the Earth.

“Galactic Watering Hole”

The new method calls for a different approach. Instead of looking for these straight lines, the researchers entered original observations and then simulated the type of signals they were interested in and trained their algorithm to recognize these signals.

This allows for a more flexible approach to signal recognition, picking up narrowband anomalies that go on and off even if they don’t have the simple line shape that would mark the traditional algorithm. It’s a more general approach, allowing for the possibility of signal types that may not be predicted by Earth-based engineers.

It also makes the approach faster and more efficient. “What people originally did is they took classical algorithms and added a machine learning approach somewhere in the pipeline,” Ma said. Now, with developments in machine learning, the whole pipeline can be based on machine learning.

This is important, as SETI is essentially a numbers game: the challenge is getting enough data from enough telescopes to increase the probability of a detection. Sifting through all that data to find the needle in the cosmic haystack requires increasingly efficient methods.

SETI is essentially a numbers game

The wider field of SETI is an unusual endeavor because researchers can devote their entire careers to looking for something that may or may not be there. New tools such as powerful telescope arrays and machine learning techniques can help make the search more precise and accurate. “There’s never been a better time in history to do SETI,” Ma said. But even if life exists beyond our planet, we may never have the chance to discover it.

On the other hand, there’s the tantalizing possibility that humanity could detect an intriguing signal tomorrow or even that evidence of an alien civilization might already exist in the masses of data gathered from decades of listening to the skies.

That prospect of satisfying their curiosity keeps SETI researchers on their long hunt. “Who knows. Maybe there’s a groundbreaking signal on a hard drive in a basement right now,” Ma said. “Someone’s got to look, right?”

Leave a Reply

Your email address will not be published. Required fields are marked *