Astronomers and computer scientists from the University of Warwick have confirmed the discovery of 50 new planets.
The fascinating discovery was made using artificial intelligence, and confirmed by a new machine learning algorithm developed by scientists at the university’s astronomy department.
The size of the discovered planets range from worlds as large as Neptune to those that are smaller than the Earth.
It is the first time that astronomers have used artificial intelligence to analyse a sample of potential planets to determine which ones are real and and which are ‘fakes’.
The artificial intelligence calculated the probability of each candidate to be a true planet.
By confirming that these 50 planets are real, astronomers can now prioritise these for further observations with dedicated telescopes.
You can get your Coventry and Warwickshire news straight to your email inbox – and it’s FREE!
All you need to do is sign up here.
The emails come out twice a day, at lunchtime and in the evening, with the latest news, what’s on and sport from across Coventry and Warwickshire.
Sky surveys had found thousands of planet candidates, and astronomers had to separate the true planets from fake ones.
This new system will make for more efficient planet identification, as the old way of confirming the existence of new planets would require a search through huge amounts of data from telescopes for the signs of planets passing between the telescope and their star, known as transiting.
The results could be effected by even a slight error in the camera, or a dip in light from the star that the telescope detects, which in turn could lead to a ‘false positive’.
Now, the new algorithm designed by the experts at the University of Warwick is faster than previous techniques, can be automated, and improved with further training.
Dr David Armstrong, from the University of Warwick Department of Physics, said: “The algorithm we have developed lets us take fifty candidates across the threshold for planet validation, upgrading them to real planets.”
He added: “In terms of planet validation, no-one has used a machine learning technique before.”
The discovery may also have a large impact on the research field in general, because once built and trained the algorithm is faster than existing techniques and can be completely automated, making it ideal for analysing the potentially thousands of planetary candidates observed in current surveys like TESS (Transiting Exoplanet Survey Satellite).
The work that led to this discovery was done by researchers from Warwick’s Departments of Physics and Computer Science, as well as Turing fellows from The Alan Turing Institute.
The researchers argue that it should be one of the tools to be collectively used to validate planets in future.
So what now, can we expect to see further discoveries? Dr Armstrong said: “Almost 30% of the known planets to date have been validated using just one method, and that’s not ideal. Developing new methods for validation is desirable for that reason alone. But machine learning also lets us do it very quickly and prioritise candidates much faster.
“We still have to spend time training the algorithm, but once that is done it becomes much easier to apply it to future candidates. You can also incorporate new discoveries to progressively improve it.”