topbanner_forum
  *

avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • Thursday October 10, 2024, 11:24 am
  • Proudly celebrating 15+ years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Author Topic: IBM develops a computer chip with one million 'neurons' that 'functions like a h  (Read 5265 times)

Renegade

  • Charter Member
  • Joined in 2005
  • ***
  • Posts: 13,291
  • Tell me something you don't know...
    • View Profile
    • Renegade Minds
    • Donate to Member
Embrace your infant overlords! :P

http://www.dailymail...ike-human-brain.html

  • TrueNorth is being hailed as the world’s first neurosynaptic computer chip because it can figure things out on its own
  • Modern processors have 1.4 bn transistors and consume up to 140 watts but the IBM chip contains 5.4 bn transistors and uses just 70 milliwatts
  • Richard Doherty, the research director of tech research firm Envisioneering Group, hailed IBM's chip as a ‘really big deal’

IBM has developed a computer chip which it says will function like a human brain in a giant step forward for artificial intelligence.

TrueNorth is being hailed as the world’s first neurosynaptic computer chip because it can figure things out on its own.

The chip also has one million ‘neurons’ and could cram the same power as a super computer into a circuit the size of a postage stamp.

IBM's TrueNorth processor could enable a wide variety of applications based on the human brain's computing power. For instance, it could help assist vision-impaired people to navigate through an environment

Experts said that it was as big an advance as the advent of supercomputers in the 1980.

Horst Simon, deputy director of the Lawrence Berkeley National Laboratory, told the New York Times: ‘It is a remarkable achievement in terms of scalability and low power consumption’.

Modern processors have some 1.4 billion transistors and consume up to 140 watts but the IBM chip contains 5.4 billion transistors and uses just 70 milliwatts of power, meaning it is incredibly efficient.

More about your new masters at the link! :P
Slow Down Music - Where I commit thought crimes...

Freedom is the right to be wrong, not the right to do wrong. - John Diefenbaker

kunkel321

  • Supporting Member
  • Joined in 2009
  • **
  • Posts: 599
    • View Profile
    • Donate to Member
Hopefully they can combine this new technology with those self-replicating 3D printers...   :P

Renegade

  • Charter Member
  • Joined in 2005
  • ***
  • Posts: 13,291
  • Tell me something you don't know...
    • View Profile
    • Renegade Minds
    • Donate to Member
Hopefully they can combine this new technology with those self-replicating 3D printers...   :P

You mean like Replicators from Star Gate? :D
Slow Down Music - Where I commit thought crimes...

Freedom is the right to be wrong, not the right to do wrong. - John Diefenbaker

MilesAhead

  • Supporting Member
  • Joined in 2009
  • **
  • Posts: 7,736
    • View Profile
    • Donate to Member
It would've been funnier if it said "Richard Doherty, the former research director of tech research firm Envisioneering Group(he has since been replaced by the prototype) said..."

MilesAhead

  • Supporting Member
  • Joined in 2009
  • **
  • Posts: 7,736
    • View Profile
    • Donate to Member
Hmm, if a chip has more than a million neurons does that make it a moron?

SeraphimLabs

  • Participant
  • Joined in 2012
  • *
  • Posts: 497
  • Be Ready
    • View Profile
    • SeraphimLabs
    • Donate to Member
Just being able to create synaptic mesh networks does not make a functional intelligence.

Although we know what the human gray matter is made of, we don't have anywhere near a complete mapping of its connectivity to use as a template for intelligences.

Developing artificial intelligence using these new synaptic chips is going to be a bit of a guessing game for quite some time until someone finally stumbles into it.

However, I see one very radical application right off the top- Prosthetics.

Right now we are just starting to be able to interface the human body directly to electronic constructs and use that to control motion in a natural manner.

If you used a synaptic design as the computing device in such a prosthetic device, it would be possible for the device to learn the user's behavior and adjust its output accordingly to be as lifelike and natural-fitting as their original limb would have been.

This also opens up the possiblity of more and more sophisticated prosthetics, replacing even sections of the eyes and ears or perhaps the entire lower body by using the synaptic learning ability to train the prosthetic to respond to its wearer's thoughts and neural impulses.

It seems that the age of cybernetics is about to begin at last.

Renegade

  • Charter Member
  • Joined in 2005
  • ***
  • Posts: 13,291
  • Tell me something you don't know...
    • View Profile
    • Renegade Minds
    • Donate to Member
Just being able to create synaptic mesh networks does not make a functional intelligence.

Although we know what the human gray matter is made of, we don't have anywhere near a complete mapping of its connectivity to use as a template for intelligences.

Developing artificial intelligence using these new synaptic chips is going to be a bit of a guessing game for quite some time until someone finally stumbles into it.

It's a fascinating field. Roger Penrose argued in "The Emperor's New Mind" that AI was a pipe dream in the sense of actual consciousness.

But, who knows?
Slow Down Music - Where I commit thought crimes...

Freedom is the right to be wrong, not the right to do wrong. - John Diefenbaker

MilesAhead

  • Supporting Member
  • Joined in 2009
  • **
  • Posts: 7,736
    • View Profile
    • Donate to Member
Strosser used this quote in the beginning of Accelerando

The question of whether Machines Can Think... is about as relevant as the question of whether Submarines Can Swim.

    Dijkstra (1984) The threats to computing science (EWD898)

I never thought of it quite that way.  :)

I have about 70 pages left to read.


kunkel321

  • Supporting Member
  • Joined in 2009
  • **
  • Posts: 599
    • View Profile
    • Donate to Member
You mean like Replicators from Star Gate? :D
  Yep.  That's where I was headed!   :Thmbsup: