Scientists Developed New Method ‘AI’ To Locate Planets Far Away From Earth

(CTN News) – An innovative method for discovering and classifying new planets that are far from Earth has been developed by a team of scientists from the University of Georgia.

Artificial Intelligence (AI) Extension

An exoplanet using machine learning, an advancement of artificial intelligence (AI), has been discovered.

A recent study showed that by scanning protoplanetary discs, the gas surrounding young stars, AI can accurately determine the presence of the exoplanet.

The recently released results represent a first step in using machine learning to identify previously ignored exoplanets.

In a statement, Jason Terry, the study’s lead author, said: “We confirmed the planet using conventional methods, but our models drove us to do those simulations and showed us exactly where the planet would lie.

“Despite having already been examined, our models discovered a disc when we applied them to a collection of older observations. We performed simulations of the disc and discovered that a planet might replicate the finding, similar to earlier results.

Several images that sharply focused on a specific region of the disc that turned out to have the typical indication of a planet – an extraordinary variation in the velocity of the gas around the planet – indicated the existence of a planet, according to Terry.

Machine Learning Used To Locate Known-Forming Exoplanets

This is an extraordinarily exciting proof of concept, according to Cassandra Hall, an assistant professor of computational astrophysics and the leader of the Exoplanet and Planet Formation Research Group at the University of Georgia.

From our earlier research, we were aware that machine learning could be used to locate known-forming exoplanets. We can utilise it to discover new brands with certainty now that we are aware of this.

This shows that our models, and machine learning in general, are capable of swiftly and precisely identifying crucial information that humans may overlook.

This might significantly speed up processing and ensuing theoretical insights, according to Terry.

“We think there will be an important place for these types of techniques as our datasets get even larger,” the researchers said of their analysis of the full catalogue, which only took around an hour.