The Kepler-90 system is similar to our own with 8 planets, but is not suspected of harboring any life.
Vanderburg, a NASA Sagan fellow at UT Austin, and Shallue, a Google machine learning researcher, teamed up to train a computer to learn how to identify signs of an exoplanet in the light readings from distant stars recorded by Kepler. Additionally, this planet orbits its star every 14 days, which is quite faster than Earth's 365 days.
It means that life as we know it has no chance of existing, with the planet's average surface temperature believed to exceed 800F (426C). In fact, all eight planets are scrunched up around this star, orbiting closer than Earth does to our sun.
The finding sets a new high for the most exoplanets found orbiting just one star and makes the planetary system a competitor of the solar system in terms of the total number of known planets.
As explained in a press release from NASA, researchers were able to identify the previously missed planet by adopting machine learning techniques that are created to find patterns in data the same way human brains do. This shows that there really are benefits to employing machine learning as a technique for searching the cosmos for interesting research subjects without the need for a human to slog through thousands of signals in order to find a few interesting stars that warrant a closer look. Over the past four years, Kepler has detected over 35,000 possible planetary signals, with rudimentary automated tests and even human eyes serving as the only vetting tools available to mine the information. The discovery of Kepler-90i is notable because it's the eighth planet circling the Kepler-90 star. The algorithm eventually achieved a 96 percent success rate in identifying which signals reflected planets and which did not. First, they trained the neural network to identify transiting exoplanets in a set of 15,000 previously vetted signals from the Kepler exoplanet catalog.
"We got lots of false positives of planets but also potentially more real planets", Vanderburg said.
Kepler-90i was not the only jewel this neural network sifted out.
The neural network also discovered one other planet, Kepler-80g. The result is an extremely stable system, similar to the seven planets in the TRAPPIST-1 system, so precisely balanced that the length of Kepler-80g's year could be predicted with mathematics.