How TensorFlow Helped Discover 2 Planets

by Developer.com Staff

NASA used Google's artificial intelligence software to examine starlight.

This week, NASA announced that it had discovered another solar system that has eight planets, just like our own. The story behind the story is that the space agency was able to make the discovery thanks to Google's TensorFlow open source machine learning software.

An article at eWeek explains how Google AI researcher Chris Shallue and Andrew Vanderburg, an astrophysicist at the University of Texas, Austin, used TensorFlow to analyze starlight data from the Kepler space telescope. The system was able to identify times when the starlight dimmed because two newly discovered exoplanets were passing in front of the stars. That led directly to the discovery of the planets.

View article

This article was originally published on Friday Dec 15th 2017
Mobile Site | Full Site