In the 17th century, when Galileo first spied the moon's mountains, rings encircling Saturn, and satellites orbiting Jupiter, an early telescope of his own making aided his eye. Now astronomers are adding a new tool to their arsenal.

A team of researchers led by David Armstrong at the University of Warwick in the UK recently trained a machine-learning algorithm to identify "exoplanets"—that is, planets outside our solar system—from NASA data. The team used the tool to confirm 50 new potential planets, a first for artificial intelligence as applied to astronomy.

"Our models can validate thousands of unseen candidates in seconds," the study's authors wrote in the abstract to their paper, which appears in the Monthly Notices of the Royal Astronomical Society. Given the gargantuan size of many astronomical datasets, the method could greatly boost the speed of discovery for world-hunting.

Scientists, including ones employed by Google, previously used machine learning to identify possible exoplanets. The new experiment represents, however, the first time scientists have applied machine learning to "validation," a further step toward confirming results that involves additional statistical calculation.

The trick is to separate real planets from fake-outs. "Rather than saying which candidates are more likely to be planets, we can now say what the precise statistical likelihood is," Armstrong said in a statement. "Where there is less than a 1% chance of a candidate being a false positive, it is considered a validated planet."

The planets plucked out by the scientists' program range from Neptune-size to smaller than Earth, the researchers said. The duration of their sojourns around their respective stars last anywhere from 200 days to as little as a single day.

Astronomers seek out potential exoplanets by looking for fluctuations in the brightness of distant stars. Periodic dimming of starlight could indicate the presence of orbiting passersby, like planets. But aberrations and other celestial objects, like asteroids, can be deceptive.

The researchers trained their algorithm on already-parsed datasets collected by NASA's retired, planet-hunting Kepler mission. Once the algorithm got the hang of telling confirmed planets from false positives, the scientists fed it data containing as yet unconfirmed planetary candidates. The result: 50 new planets passed muster.

The A.I. technique could be applied to data collected by other space probes. That includes NASA's Transiting Exoplanet Survey Satellite, launched in April 2018. The team behind that telescope wrapped up its primary mission, a two year-long survey, this summer after finding 66 new exoplanets and 2,100 candidate exoplanets.

The scientists say their tool can work with TESS as well as the European Space Agency's upcoming PLATO mission, which stands for "planetary transits and oscillations of stars." The authors note "the large number of observed targets necessitates the use of automated algorithms."

The University of Warwick's Theo Damoulas, a computer science professor and another author of the paper, said that A.I. techniques "are especially suited for an exciting problem like this in astrophysics."