An Artificial Intelligence (AI) and Machine Learning (ML) based model developed by Google that helped the NASA space agency to discover two exoplanets last December, has now been officially rolled out publicly for researchers to use.
The announcement was made by Chris Shallue, a senior software engineer at Google Brain Team, in a blog post last week. In the blog post, Shallue explained how they trained their neural network to analyze data from NASA’s Kepler space telescope and pick out the signals that present the most promise of being exoplanets. While the analysis only covered about 700 stars, it was enough to prove that Google’s AI model is helpful in identifying exoplanets using the Kepler data.
The blog post further explained how scientists use automated software to detect signals that might be caused by planets, and then manually follow up to decide whether each signal is a planet or a false positive. Notably, during the course of finding the two exoplanets, scientists manually reviewed more than 30,000 different signals gleaned from the Kepler data and approximately 2,500 all those signals turned out to be planets.
The code for the exoplanet-hunting algorithm and instructions on how to use it has been released on Github. While the code, as well as the Kepler data, is available for use by anyone, it’s not exactly ‘plug and play.’ Aspiring exoplanet hunters who have experience of using Google’s machine learning software TensorFlow and coding in Python will be able to use the exoplanet-hunting algorithm in a better way.
However, Shallue believes that releasing the code will encourage further analysis of the Kepler data. He is also hopeful that the neural net will pave the way for still more sophisticated exoplanet hunting tools in the future.
“We hope this release will prove a useful starting point for developing similar models for other NASA missions, like K2 (Kepler’s second mission) and the upcoming Transiting Exoplanet Survey Satellite mission,” Shallue noted in the blog post.