More than 300 exoplanets added to list, thanks to a new deep learning method 


More than 300 exoplanets have been discovered in deep space thanks to a newly created algorithm using data from NASA’s spacecraft and supercomputer

  • An additional 301 exoplanets have been confirmed, thanks to a new deep learning algorithm
  • The ExoMiner deep neural network was created using data from NASA’s Kepler spacecraft and its supercomputer, Pleiades
  •  The newly confirmed planets brings the total of confirmed exoplanets to 4,870


An additional 301 exoplanets have been confirmed, thanks to a new deep learning algorithm, NASA said. 

The significant addition to the ledger was made possible by the ExoMiner deep neural network, which was created using data from NASA’s Kepler spacecraft and its follow-on, K2.

It uses the space agency’s supercomputer, Pleiades and is capable of deciphering the difference between real exoplanets and ‘false positives.’

The newly confirmed planets, which orbit distant stars in the universe, brings the total of confirmed exoplanets to 4,870. 

An additional 301 exoplanets have been confirmed, thanks to a new deep learning algorithm

An additional 301 exoplanets have been confirmed, thanks to a new deep learning algorithm

The newly confirmed planets brings the total of confirmed exoplanets to 4,870

The newly confirmed planets brings the total of confirmed exoplanets to 4,870

‘Unlike other exoplanet-detecting machine learning programs, ExoMiner isn’t a black box – there is no mystery as to why it decides something is a planet or not,’ one of the study’s authors, Jon Jenkins, exoplanet scientist at NASA’s Ames Research Center in a statement. 

‘We can easily explain which features in the data lead ExoMiner to reject or confirm a planet.’

There is a slight difference between a ‘confirmed’ and a ‘validated’ exoplanet, NASA notes: exoplanets are ‘confirmed’ when different observation techniques highlight features exhibited only by planets; they are ‘validated’ when statistics are used. 

In the study, ExoMiner used data sets from the Kepler Archive to discover the 301 planets from a much larger set of candidates.

The ExoMiner deep neural network was created using data from NASA's Kepler spacecraft and its supercomputer, Pleiades (pictured)

The ExoMiner deep neural network was created using data from NASA’s Kepler spacecraft and its supercomputer, Pleiades (pictured)

The ExoMiner deep neural network was created using data from NASA's Kepler spacecraft (pictured) and its supercomputer, Pleiades

The ExoMiner deep neural network was created using data from NASA’s Kepler spacecraft (pictured) and its supercomputer, Pleiades

They were validated by the Kepler Science Operations Center pipeline and promoted to planet candidate status by the Kepler Science Office.

The newly published research shows that the neural network is more consistent and precise when it gets rid of false positives than human scientists.

It also gives the researchers additional detail into why ExoMiner made the decision that it did.

‘When ExoMiner says something is a planet, you can be sure it’s a planet,’ added Hamed Valizadegan, ExoMiner project lead and machine learning manager. 

‘ExoMiner is highly accurate and in some ways more reliable than both existing machine classifiers and the human experts it’s meant to emulate because of the biases that come with human labeling.’ 

Of the 301 exoplanets that were added to the ever-growing list, none are ‘ believed to be Earth-like or in the habitable zone of their parent stars,’ NASA notes, but some of them share certain characteristics of other exoplanets near Earth.

‘These 301 discoveries help us better understand planets and solar systems beyond our own, and what makes ours so unique,’ said Jenkins. 

The research was recently published in the Astrophysical Journal. 

Advertisement

Leave a Reply

Your email address will not be published.