Thanks To AI More Than 300 Exoplanets Successfully Revealed, Habitable?

More than 300 exoplanets have been discovered by NASA scientists. (photo: doc. unsplash)

JAKARTA - More than 300 exoplanets have been discovered by NASA scientists from the catalog of the now-defunct telescope, the Kepler Space Telescope.

The Kepler Space Telescope is NASA's first telescope to work on planet hunting. He has observed hundreds of thousands of stars in the search for potentially habitable worlds outside the Solar System.

Although the telescope is no longer functional, a new artificial intelligence algorithm has found more than 300 previously unknown exoplanets in the data collected by the telescope. This research has been published in the Astrophysical Journal.

In the catalog of potential planets that the Kepler telescope has compiled, it continues to produce new discoveries even after the telescope dies. Scientists analyze data for signs of exoplanets. However, a new algorithm called ExoMiner can now emulate that procedure and explore the catalog more quickly and efficiently.

Citing Soace, Monday, November 29, the telescope, known to stop working in November 2018, was looking for a decrease in star brightness that might be caused by a planet passing in front of the stellar disk, as seen from Kepler's point of view.

But not all of the brightness dips are caused by exoplanets, and scientists have to follow a complicated procedure to distinguish false positive signals from real ones.

ExoMiner is a so-called neural network, a type of artificial intelligence algorithm that can learn and improve its capabilities when entering a sufficient amount of data. And Kepler generates a lot of data.

In just under 10 years of service, the telescope was able to find thousands of planet candidates, nearly 3,000 of which have been confirmed. That is the bulk of the total 4,569 exoplanets known today.

For each exoplanet candidate, scientists looking at the Kepler data will look at the light curve and calculate how much of the star the planet appears to be covering.

They will also analyze how long it takes the would-be planet to cross the stellar disk. In some cases, the observed brightness changes are unlikely to be explained by the orbiting exoplanet.

ExoMiner's algorithm follows exactly the same but more efficient process, allowing scientists to add 301 previously unknown exoplanets to Kepler's planetary catalog at once.

"When ExoMiner says something is a planet, you can be sure it's a planet," said Hamed Valizadegan, ExoMiner project leader.

"ExoMiner is highly accurate and in some ways more reliable than existing machine classifiers and human experts it is meant to emulate due to the biases that come with human labeling."

Now that ExoMiner is proving its worth, scientists want to use it to help sift through data from existing and future exoplanet-search missions, such as NASA's current Transiting Exoplanet Survey Satellite (TESS) or the European Space Agency's (PLATO) Planetary Transits and Oscillations of Stars. , a mission to be launched in 2026.

Unfortunately, none of the newly confirmed exoplanets are likely candidates to host life, because they are outside the habitable zone of their parent star.