• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
  LongeCity
              Advocacy & Research for Unlimited Lifespans


Adverts help to support the work of this non-profit organisation. To go ad-free join as a Member.


Photo

Computer Program Self-Discovers Laws of Physics


  • Please log in to reply
9 replies to this topic

#1 Infinitytimes2

  • Guest
  • 31 posts
  • 1
  • Location:Nowhere New Jersey

Posted 03 April 2009 - 06:29 AM


http://blog.wired.co...4/newtonai.html

Can anyone imagine it possible to develop a sort of engine based environment on a super computer taking in to account the laws of physics for the simulation of genetic regulatory systems, with software that has algorithms specially designed to measure all the developmental process variables we face in the biological world and perhaps with a genome life can be simulated starting from the first cell division and passed embryonic stages?

is this exaflop talk? or are we almost there?

Edited by Infinitytimes2, 03 April 2009 - 07:18 AM.


#2 modelcadet

  • Guest
  • 443 posts
  • 7

Posted 03 April 2009 - 03:48 PM

http://blog.wired.co...4/newtonai.html

Can anyone imagine it possible to develop a sort of engine based environment on a super computer taking in to account the laws of physics for the simulation of genetic regulatory systems, with software that has algorithms specially designed to measure all the developmental process variables we face in the biological world and perhaps with a genome life can be simulated starting from the first cell division and passed embryonic stages?

is this exaflop talk? or are we almost there?


Thanks for the post, doubleinfinity. Combine that with this:
http://www.physorg.c...s157900674.html

Scientists have created a Robot Scientist which the researchers believe is the first machine to have independently discovered new scientific knowledge. The robot, called Adam, is a computer system that fully automates the scientific process. The work will be published tomorrow (03 April 2009) in the journal Science.



sponsored ad

  • Advert

#3 Infinitytimes2

  • Topic Starter
  • Guest
  • 31 posts
  • 1
  • Location:Nowhere New Jersey

Posted 03 April 2009 - 04:36 PM

It's almost scary, this new era were entering. I'd like to hear some of imminst opinions on safety precaution and protocols that should be taken to make sure this technology doesn't get out of hand.

#4 Putz

  • Guest, F@H
  • 55 posts
  • 0
  • Location:Providence, RI

Posted 04 April 2009 - 01:46 AM

The Singularity Is Near :~

#5 valkyrie_ice

  • Guest
  • 837 posts
  • 142
  • Location:Monteagle, TN

Posted 04 April 2009 - 02:50 AM

welcome to the world of brute force attacks on data mining with the added ability to perform physical experiments.

What this makes likely is a rapid increase in the development cycles of scientific discoveries due to the automation of simple time consuming brute force approaches to such fields as chemical synthesis and other fields in which trial and error play a significant role in discoveries. It will likely take five years before we will see this in regular use to run massively parallel batch processing experiments, but when we do, I'd predict a rapid fallout effect.

The next decade seems to promise massive change.

#6 niner

  • Guest
  • 16,276 posts
  • 2,000
  • Location:Philadelphia

Posted 04 April 2009 - 06:24 AM

The next decade seems to promise massive change.

That it does, although not necessarily because of this particular case, which is a bit overhyped, IMHO. As Connor McLeod mentioned in the other thread on this paper, it's just a data mining approach that relies on large quantities of data. To characterize it as AI is somewhat misleading. The next decade will still blow your doors off, for various reasons.

#7 valkyrie_ice

  • Guest
  • 837 posts
  • 142
  • Location:Monteagle, TN

Posted 04 April 2009 - 09:25 AM

The next decade seems to promise massive change.

That it does, although not necessarily because of this particular case, which is a bit overhyped, IMHO. As Connor McLeod mentioned in the other thread on this paper, it's just a data mining approach that relies on large quantities of data. To characterize it as AI is somewhat misleading. The next decade will still blow your doors off, for various reasons.



Yes, but it is automated data mining with robots. set up in massively parallel mode, they can trial and error test many scientific processes automatically. could speed things up a lot.

#8 treonsverdery

  • Guest
  • 1,315 posts
  • 160
  • Location:where I am at

Posted 21 June 2011 - 09:57 PM

it would be beneficial to make the query language preference based, otherwise the cliche comic observer creates universe could occur with a nonsentient observer. actually with the notion of combinatorial emergent properties some perceptually possible forms of observation change their environment. Rather than say "what is it", the query language should say, does it permit levitation, or I suppose among the brave describe why the universe permits levitation.

say, have any of you noticed the internet is very pleasant?

Edited by treonsverdery, 21 June 2011 - 09:58 PM.


#9 Shoe

  • Guest, F@H
  • 135 posts
  • 1

Posted 21 June 2011 - 10:59 PM

Spambot alert.

sponsored ad

  • Advert

#10 DAMABO

  • Guest
  • 181 posts
  • 4
  • Location:Mars

Posted 12 April 2012 - 09:54 PM

Spambot alert.


you're such a cool guy. love the combination of your username and your photo ;) .




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users