It looks like you're using an Ad Blocker.

Please white-list or disable AboveTopSecret.com in your ad-blocking tool.

Thank you.

 

Some features of ATS will be disabled while you continue to use an ad-blocker.

 

Google teaching computers to mimic human brain

page: 1
3

log in

join
share:

posted on Jul, 18 2012 @ 05:33 AM
link   

Computers programmed with algorithms intended to mimic neural connections "learned" to recognise cats after being shown a sampling of YouTube videos, Google fellow Jeff Dean and visiting faculty Andrew Ng said in a blog post.

"Our hypothesis was that it would learn to recognise common objects in those videos," the researchers said.

"Indeed, to our amusement, one of our artificial neurons learned to respond strongly to pictures of... cats," they continued.

Google teaching computers to mimic human brain

Very interesting stuff. It seems like we are only just beginning to tap into the power of self-learning artificial neural networks, even though the technology has existed for over a decade and the potential of these systems has been understood and demonstrated for over a decade. It makes me wonder how far the black budget projects have developed this technology... because I would be willing to bet a million dollars on the fact that they have developed it beyond anything we see in the public realm of technology.
edit on 18/7/2012 by ChaoticOrder because: (no reason given)



posted on Jul, 18 2012 @ 05:54 AM
link   
Amazing but scary.

What's the true purpose of it, how will it be used?



posted on Jul, 18 2012 @ 05:59 AM
link   
reply to post by violet
 


Why is it scary? I find the entire concept fascinating.

Though I do agree that they have had artificial neural networks in the past with the ability to detect specific pattens out of an image, so it's not exactly new, but that was then, this is now - computers have come a long way since then, and I can only imagine what, with enough funding and the right brain behind the technology, we will see in the next decade or so..




posted on Jul, 18 2012 @ 06:02 AM
link   
reply to post by violet
 



What's the true purpose of it, how will it be used?
You mean what it's Googles true motive? No idea... perhaps to spot copyrighted material in the videos, although it already seems to me as if they do that. I uploaded a long documentary video which I had made myself only a few weeks ago, and within literally minutes after I had uploaded it, I had multiple copyright notices. I don't see how they could have known so quickly unless they scanned the video with some sort of advanced recognition technology.

But in the general realm of possibilities, this technology has a potentially infinite number of purposes. There are so many things which we can't easily define and quantize in a way which makes it easy for computers to understand. This technology behaves in a way such that it can interpret and understand very abstract concepts and messy real world logic, which could be extremely useful in some many different areas.



posted on Jul, 18 2012 @ 06:52 AM
link   

Originally posted by violet
Amazing but scary.

What's the true purpose of it, how will it be used?


One huge possibility I can think of is sorting through flagged conversations and looking for anything suspicious. Remember that list of forbidden words? Maybe someone on here says the word "shooting" in relation to a common, everyday news article... Or maybe they're talking about how they're about to go shoot up their local school. Now I'm not saying I support or do not support surveillance (that's beyond the point), but technology like this could potentially make it far easier to monitor such conversations.



posted on Jul, 18 2012 @ 07:12 AM
link   
I love technology, especially when it comes to artificial intelligence, but you just know that you can't advance technogy without the Government weaponising it, especially with robotics and computer technology.

Just watch Penn and Teller: Tell A Lie. In the episode where Teller gets into a tank with pirhana, there is a 5 minute section where there is a little quadrocopter robot that can fly around without any humans controlling it, catch a ball and lift objects, create it's own 3D map of the vicinity it is in, use that map to avoid obstacles, carry out commands, and many other things, however, would you like something as advanced to become weaponised? Of course not, and it is obvious that this technology would be more "advantageous" to the Government and it's military operations, rather than using it for things such as surveying dangerous buildings, going down mine-shafts etc.


 
Posted Via ATS Mobile: m.abovetopsecret.com
 




top topics
 
3

log in

join