United Kingdom (U.K.) researchers used a highly-advanced machine learning algorithm to identify 50 new planets based on NASA's old data. The recent discovery of British scientists marked a technological breakthrough in astronomy.
AI confirms 50 new planets from old NASA data, in a groundbreaking first https://t.co/1NqKpuyzT1 #MachineLearning #DeepLearning pic.twitter.com/yWd9X0Fw2X — Paul Lopez (@lopezunwired) August 25, 2020
The University of Warwick's computer scientists and astronomers developed a machine-learning algorithm to understand old NASA data, which contains thousands of possible planet candidates. However, experts find it hard to identify the genuine candidates.
Also Read: DARPA's AlphaDogfight Trial's Final Round: Human F-16 Pilot Lost Against AI For the First Time
Space scientists and researchers look for flashes of light that could indicate a planet is passing between the telescope and a star whenever they are searching for exoplanets. However, the dips of lights are still not reliable since they could be caused by other factors such as errors in the camera or background interference.
A new artificial intelligence (AI) was developed to solve this issue. The researchers used the new algorithm in the data collected by NASA's now-retired Kepler Space Telescope to further enhance its ability.
Kepler spent nine years in deep space on a world-hunting mission. The machine-learning algorithm was used to analyze old data sets that had not yet been confirmed after it learned to separate false planets from real ones, accurately.
Other things the new technology can be used to
The scientists identified the 50 exoplanets orbiting around other stars. University explained in a news release that their sizes range from Neptune to smaller the Earth.
Breakthrough AI identifies 50 new planets from old NASA data https://t.co/Jidl33VxSL pic.twitter.com/mlV7hMUVC8 — Steven (@Boog1e_Man831) August 26, 2020
Some of the planets' orbits are as short as a single day, and others are as long as 200. The scientists will prioritize the identified new worlds after they confirmed that they're real.
The Monthly Notices of the Royal Astronomical Society published the study's findings last week.
"In terms of planet validation, no-one has used a machine learning technique before," said David Armstrong, a University of Warwick's scientists and one of the study's lead authors.
"Machine learning has been used for ranking planetary candidates but never in a probabilistic framework, which is what you need to truly validate a planet," he added.
For more news updates about space studies, always keep your tabs open here at TechTimes.
This article is owned by TechTimes,
Written by: Giuliano de Leon.
