Google's Machine Learning Is Analyzing Data From NASA's Kepler Space Telescope (nasa.gov) 27
NASA writes:
NASA will host a media teleconference at 1 p.m. EST Thursday, Dec. 14, to announce the latest discovery made by its planet-hunting Kepler space telescope. The discovery was made by researchers using machine learning from Google. Machine learning is an approach to artificial intelligence, and demonstrates new ways of analyzing Kepler data... When Kepler launched in March 2009, scientists didn't know how common planets were beyond our solar system. Thanks to Kepler's treasure trove of discoveries, astronomers now believe there may be at least one planet orbiting every star in the sky.
Space.com adds: Kepler spots alien worlds by noticing the tiny brightness dips they cause when they cross the face of their host star from the spacecraft's perspective. Kepler is the most accomplished planet hunter in history. It has found more than 2,500 confirmed alien worlds -- about 70 percent of all known exoplanets -- along with a roughly equal number of "candidates" that await confirmation by follow-up observations or analyses. The vast majority of these discoveries have come via observations that Kepler made during its original mission, which ran from 2009 to 2013. Study of these data sets is ongoing; over the past few years, researchers have used improved analysis techniques to spot many exoplanets in data that Kepler gathered a half-decade ago or more.
Space.com describes Thursday's announcement as an exoplanet discovery. (Earlier they reported on the discovery of "a possibly habitable alien world" about 2.2 times the size of earth orbiting a dwarf star "within the range of distances where liquid water could exist on a world's surface".)
Slashdot reader schwit1 points out that other less-credible sites speculate NASA's announcement will be "a major discovery about life beyond earth."
Space.com adds: Kepler spots alien worlds by noticing the tiny brightness dips they cause when they cross the face of their host star from the spacecraft's perspective. Kepler is the most accomplished planet hunter in history. It has found more than 2,500 confirmed alien worlds -- about 70 percent of all known exoplanets -- along with a roughly equal number of "candidates" that await confirmation by follow-up observations or analyses. The vast majority of these discoveries have come via observations that Kepler made during its original mission, which ran from 2009 to 2013. Study of these data sets is ongoing; over the past few years, researchers have used improved analysis techniques to spot many exoplanets in data that Kepler gathered a half-decade ago or more.
Space.com describes Thursday's announcement as an exoplanet discovery. (Earlier they reported on the discovery of "a possibly habitable alien world" about 2.2 times the size of earth orbiting a dwarf star "within the range of distances where liquid water could exist on a world's surface".)
Slashdot reader schwit1 points out that other less-credible sites speculate NASA's announcement will be "a major discovery about life beyond earth."
Planet hunting is nice- (Score:2)
But I'd be a little happier if it was asteroid-hunting. Particularly asteroids that look as though they might hit Earth any time soon.
Just sayin'... while you've got the two devices connected together...
Re: (Score:2)
But I'd be a little happier if it was asteroid-hunting.
i'd be happier if nasa, and those who do real science, refrain from tagging currently fashionable buzzwords like "machine learning", "artificial intelligence", etc, straight out of corporate hype to their own revelations.
"google's data analysis helped in analysis of kepler data, in making yet another one discovery to add to others already made", would be the straight real story here
Re: (Score:1)
I'd be happier if nasa, and those who do real science,
I think Richard Feynman would object to that statement
http://www.ralentz.com/old/spa... [ralentz.com]
Re: (Score:3)
i'd be happier if nasa, and those who do real science, refrain from tagging currently fashionable buzzwords like "machine learning", "artificial intelligence"
You don't have a problem with buzzwords like "internet" or "database" ?
Re:Planet hunting is nice- (Score:5, Informative)
https://en.wikipedia.org/wiki/... [wikipedia.org]
>[Bennu] is a potential Earth impactor that is listed on the Sentry Risk Table with the third-highest rating on the Palermo Technical Impact Hazard Scale,[11] due to a 1-in-2,700 chance of impacting Earth in the late 22nd century.[12] It is the planned target of the OSIRIS-REx mission which is intended to return samples to Earth in 2023 for further study.
https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: (Score:2)
Faux-dramatic "reporting" makes me cringe. (Score:2)
Kepler is the most accomplished planet hunter in history.
How in the fuck many planet hunters are there, and how long have they been doing it???
Invention of telescope (Score:5, Interesting)
How in the fuck many planet hunters are there, and how long have they been doing it???
Basically, since the invention of the telescope, and thus the new possibility to see things that nobody could have seen before with a naked eye.
(e.g.: First new planets of our solar system discovered for the first time since antiquity ; William Herschel managing to discover Uranus. The Neptune getting discovered by maths. etc. - though Gallileo had observed it before but without realizing it was a new as-of-yet-unkown object)
Then better telescope offered the possibility to infere the presence of planets in other solar system than our (first one discovered at end of last century).
Eventually, a decade ago, NASA built Kepler, the best ever telescope at this task (because it was specifically designed to be able to fulfill this mission).
So in short : discovering new bodies in the sky has been a long history (since we could finally use something else than naked eyes).
Kepler is the most recent such tool, and as it was built for exo-planet discovery among its missions, it's the current best tool.
Re: (Score:2)
Where have I heard his already? (Score:2)
“But what do you do? You just there?” Case shrugged, put the vodka and the shuriken down on the cabinet and lit a Yeheyuan.
“I talk to my own kind.”
“But you’re the whole thing. Talk to yourself?”
“There’s others. I found one already. Series of transmissions recorded over a period of eight years, in the nineteen-seventies. ’Til there was me, natch, there was nobody to know, nobody to answer.”
“From where?”
“Centauri system.
A prime example (Score:3)
of putting machine learning to good use!
Why AC objects (Score:2)
So, given that we have Slashdot, and we have Google, and we have NASA, which part of this process would you like to see go away?
Slashdot telling us about it?
NASA telling slashdot about it?
NASA hunting for planets?
Google helping NASA with sponsor money?
Same for asteroids... (Score:2)
Probably not the type of information they would give out in a teleconference. Unless the alien life is coming here and the teleconference is from a bunker.
Re: (Score:2)
+1 funny, +1 insightful
Re: (Score:2)
We don't yet have the means to see if atmosphere of "earthlike" planet has signs of biological activity or composition favorable to life. The James Webb Space Telescope to be launched next year will be able to do that, as will other telescopes to be made in the next 20 years. So expect an announcement between say 2019 and 2040.
Since Earth had single cell life for so very long, and that even evolved more than once, likely that kind of life is common while complex multicellular organisms are rare.