‪‪NASA Kepler, Google AI Find 8th Planet In Faraway Galaxy

A sizzling hot eighth planet has been found in a far away solar system that rivals our own, resercheers said Thursday. The planet circling Kepler-90, a sun-like star 2,545 light-years from Earth discovered in 2013, was found in data recorded by NASA’s Kepler Space Telescope, the agency said in a joint announcement with Google, which provided “machine learning” to assist in the project.

The planet, the third rock from its sun like Earth, is being called Kepler-90i. The sizzling hot, rocky planet has a surface temperature around 800 degrees Fahrenheit — on par with Mercury in our solar system — and is about 30 percent larger than Earth.

Kepler-90 is only the other known eight-planet solar system like ours. It is a mini, “scrunched up version of our solar system,” orbiting much closer to its star than Earth does, NASA said. A “year” on Kepler 90-i passes in just 14 days, and because the surface is so hot, NASA said it’s unlikely that life exists there.

The discovery resulted from the teamwork of researchers Christopher Shallue, a senior software engineer with Google’s research team Google AI, and Andrew Vanderburg, a NASA Sagan Postdoctoral Fellow and astronomer at the University of Texas at Austin. They “trained” a computer to learn how to identify exoplanets in the light readings previously recorded by Kepler — the miniscule change in brightness captured when a planet passed in front of, or transited, a star, NASA said.

The four-year Kepler dataset consists of 35,000 possible planetary signals. It would have taken humans eons to analyze it. The artificial “neural network” Shallue and Vanderburg developed sifted through Kepler data and found weak transit signals from the previously-missed eighth planet orbiting Kepler-90, in the constellation Draco.

"Machine learning really shines in situations where there is so much data that humans can't search it for themselves,” Shallue said in a news release announcing the discovery.

“Just as we expected, there are exciting discoveries lurking in our archived Kepler data, waiting for the right tool or technology to unearth them,” Paul Hertz, director of NASA’s Astrophysics Division in Washington, D.C., said in the release. “This finding shows that our data will be a treasure trove available to innovative researchers for years to come.”

Automated tests, and sometimes human eyes, were used to verify the most promising signals in the data. However, the weakest signals often are missed using these methods. Shallue and Vanderburg thought there could be more interesting exoplanet discoveries faintly lurking in the data.

The researchers trained the neural network to identify transiting exoplanets — that is a planet outside our solar system — using a set of 15,000 previously vetted signals from Kepler. In the test set, the neural network correctly identified true planets and false positives 96 percent of the time.

Then, with the neural network having "learned" to detect the pattern of a transiting exoplanet, the researchers directed their model to search for weaker signals in 670 star systems that already had multiple known planets. Their assumption was that multiple-planet systems would be the best places to look for more exoplanets.

“We got lots of false positives of planets, but also potentially more real planets,” Vanderburg said. “It’s like sifting through rocks to find jewels. If you have a finer sieve then you will catch more rocks but you might catch more jewels, as well.”

Kepler-90i wasn’t the only gem Shallue and Vanderburg discovered using their machine-learning program. They also detected a sixth exoplanet missed in the Kepler-80g solar system.

Shallue and Vanderburg’s research paper reporting these findings has been accepted for publication in The Astronomical Journal. The researchers plan to apply their neural network to Kepler’s full set of more than 150,000 stars.

Kepler has produced an unprecedented data set for exoplanet hunting. After gazing at one patch of space for four years, the spacecraft now is operating on an extended mission and switches its field of view every 80 days.

“These results demonstrate the enduring value of Kepler’s mission,” said Jessie Dotson, Kepler’s project scientist at NASA’s Ames Research Center in California’s Silicon Valley. “New ways of looking at the data — such as this early-stage research to apply machine learning algorithms — promise to continue to yield significant advances in our understanding of planetary systems around other stars. I’m sure there are more firsts in the data waiting for people to find them.”

Artist’s concept of the Kepler-90 system compared with our own solar system. Credit: NASA/Ames Research Center/Wendy Stenzel