Recently, British astronomers and computer scientists have worked together to identify 50 new planets with the help of artificial intelligence. The researchers are from the University of Warwick. They built an ML algorithm which will be able to dig through NASA’s old data that contained thousands of potential planet candidates.
However, sometimes it is not clear which candidates are genuine and which are not. This is because while searching for exoplanets, researchers look for dips in light that indicates a planet between telescopes and stars. And sometimes, these dips can be caused by other factors, for instance, camera errors and background interference. But, the new AI has the potential to tell the difference.
The team of researchers has trained algorithm by helping it learn from NASA’s collected data of Kepler Space Telescope. Once the algorithm learned to differentiate planets, they were then used to analyze old data sets. From this, they found the 50 exoplanets.
The press release of the university said that these 50 exoplanets range in size from Neptune to smaller than Earth. Some of them even orbit as long as 200 days while some take a single day. Now that the astronomers are sure that these planets are real, they can further prioritize them for observation.
Last week, these findings were published in the Royal Astronomical Society in the Monthly Notices.
In the news release, David Armstrong of the University of Warwick, who was also one of the study’s lead authors, said “In terms of planet validation, no-one has used a machine learning technique before.” Adding “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.”