A team of astronomers in the United Kingdom has used a machine learning algorithm to analyze a sample of candidate exoplanets identified by NASA’s Kepler Space Telescope and determine which ones are real and which ones are false positives.
An artist’s impression of a compact three-planet system. Image credit: Sci-News.com.
“In terms of planet validation, no-one has used a machine learning technique before,” said lead author Dr. David Armstrong, an astronomer in the Department of Physics and the Centre for Exoplanets and Habitability at the University of Warwick.
“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.”
Dr. Armstrong and colleagues built a machine learning algorithm that can separate out real planets from fake ones in the large samples of thousands of candidates found by telescope missions.
It was trained to recognize real planets using two large samples of confirmed planets and false positives from Kepler.
The researchers then used the algorithm on a dataset of still unconfirmed planetary candidates from Kepler, resulting in 50 new confirmed planets and the first to be validated by machine learning.
These planets range from worlds as large as Neptune to smaller than the Earth, with orbits as long as 200 days to as little as a single day.
By confirming that they are real, the scientists can now prioritize these for further observations with dedicated telescopes.
“The algorithm we have developed lets us take 50 candidates across the threshold for planet validation, upgrading them to real planets,” Dr. Armstrong said.
“We hope to apply this technique to large samples of candidates from current and future missions like TESS and PLATO.”
“Probabilistic approaches to statistical machine learning are especially suited for an exciting problem like this in astrophysics that requires incorporation of prior knowledge and quantification of uncertainty in predictions,” said co-author Dr. Theo Damoulas, a researcher in the Department of Computer Science and the Department of Statistics at the University of Warwick and the Alan Turing Institute.
“A prime example when the additional computational complexity of probabilistic methods pays off significantly.”
The findings appear in the Monthly Notices of the Royal Astronomical Society.
_____
David J. Armstrong et al. Exoplanet Validation with Machine Learning: 50 new validated Kepler planets. MNRAS, published online August 20, 2020; doi: 10.1093/mnras/staa2498
