NASA Data Using AI Identify 50 Potentially New Planets

October 3, 2020 moulio No Comments

Recently, a team of researchers led by David Armstrong at the University of Warwick created an AI breakthrough. The team trained a machine-learning algorithm to identify planets outside our solar system called “exoplanets.” The algorithm quickly discovered 50 potential planets.

“Our models can validate thousands of unseen candidates in seconds,” the study’s authors wrote in the abstract to their paper. Due to the massive size of many astronomical data sets, this method of statistical analysis is a major breakthrough. Previously, the identification of potential planets has been an arduous task. This new approach changes everything.

“Rather than saying which candidates are more likely to be planets, we can now say what the precise statistical likelihood is,” Armstrong said.

The researchers used data sets from NASA’s planet-hunting Kepler mission to demonstrate AI’s great potential. Processing large amounts of data to make statistically valid predictions is something where AI surpasses human capability.