[ExI] nasa exoplanet announcement

spike spike66 at att.net
Thu Dec 14 18:40:03 UTC 2017


 

 

 

 

 

From: spike [mailto:spike66 at att.net] 
Sent: Thursday, December 14, 2017 10:07 AM
To: 'ExI chat list' <extropy-chat at lists.extropy.org>
Subject: nasa exoplanet announcement

 

 

>.NASA is in the process of announcing that the record has been tied: it
found a star with 8 planets.

 

https://www.nasa.gov/nasalive

 

>.Hit the Kepler tab.

 

>.They are also claiming that this one was found using a neural network
sifting thru archival data.  They didn't offer any details on that.

 

>.Wait, hold it, they just brought on a guy from Google AI to explain it.
They are in the process of saying this was done by AI.  Coooool!  spike

 

 

WOWsers, NASA just gave a mind-blowing announcement on Kepler, wooooohoooo!
They are claiming GoogleAI used the same software it claims can tell the
difference between a dog and a cat, then set that on the task of recognizing
exo-transits.  As far as I know, this is the first time AI has found an exo.

 

This is so cool: we often hear the words that aerospace guys really find
distasteful but true just the same: there isn't much point in building new
orbiting instruments when we don't have the budget to even look at most of
the data already collected and archived by instruments defunct for years or
even decades.  When I say most, I mean some instruments have had less than a
tenth of a percent of the data ever seen by bio-eyeballs.  

 

So. It looks to me like the next great wave in astronomy is to somehow
automate the process of digging thru skerjillions of MB of archived data.

 

Aside sorta: picture in your mind an archaeologist.  You see this guy
pecking away at rocks and sifting thru gravel at some remote site, sure.
But those are generally grad student slave labor doing that.  We have museum
cabinets filled with stuff they dug up decades ago, didn't know what it was,
filed it away, graduated, the stuff is still there to this day.
Archaeologists who go thru museums looking at what is already there make
huge discoveries, such as Steven Jay Gould wrote about in Wonderful Life in
1989, a book which rocked my world.  Those fossils were extracted in the
early 1900s, but Wolcott didn't know what they were so they ended up
forgotten in a museum until discovered decades later by Alberto Simonetta,
durn near by chance.  

 

Now we astronomy geeks want to know: how many Burgess Shale-magnitude
discoveries are sitting there as lonely bits on DVDs, waiting for someone or
something to look at them and figure out what it is telling us?  Today we
heard of the first case of an AI finding an exo-planet.  We don't have the
budget for arbitrarily many grad students, but we can replicate this AI
arbitrarily many times, and they never get bored, tired, drunk, hungry,
sleepy, stoned, distracted by some sexy grad student, none of that.  This is
a new day.

 

spike

 

 

 

 

 

 

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20171214/82db41e0/attachment.html>


More information about the extropy-chat mailing list