IBM Builds A Scalable Computer Chip Inspired By The Human Brain

forbes ^ | 7-8-2014 | Alex Knapp

[Citizen Zed]

“In 2011, we had a chip with one core,” Modha told me. “We have now scaled that to 4096 cores, while shrinking each core 15x by area and 100x by power.”

Each core of the chip is modeled on a simplified version of the brain’s neural architecture. The core contains 256 “neurons” (processors), 256 “axons” (memory) and 64,000 “synapses” (communications between neurons and axons). This structure is a radical departure from the von Neumann architecture that’s the basis of virtually every computer today (including the one you’re reading this on.)

It’s important to note, though, that the SynAPSE system won’t replace the computers of today – rather, they’re intended to supplement them. Modha likened them to co-processors used in high performance computers to help them crunch data faster. Or, in a more poetic turn as he continued talking to me, he called SynAPSE a “right-brained” computer compared to the “left-brained” architecture used in computers today.

“Current von Neumann machines are fast, symbolic, number-crunchers,” he said. “SynAPSE is slow, multi-sensory, and better at recognizing sensor data in real-time.”

[ShadowAce]

[SoCal Pubbie]

Did they use a liberal’s brain as a model?

[Alex Murphy]

[GeronL]

Everything ever made was inspired by someone’s brain!

[GeronL]

Let us hope not!

[TangoLimaSierra]

Did they use a liberal’s brain as a model?

No. Way too many synapses.

[Jewbacca]

"Did they use a liberal’s brain as a model?"

No, but liberal's brains were used to model this, an artificial liberal mind:



It's also known as a "bell jar vacuum."

[huldah1776]

Wow, my computer will finally understand when I call it a piece of @#$%#$#&& junk.

[GrandJediMasterYoda]

They’re getting closer to AI, but the problem with that though is take a look at the human brain: Half the people on this planet are completely out of their minds. So if you create an artificial intelligence equal to that of the human brain, there’s a 50% chance that it is going to think Obama is God, Muslim terrorists are freedom fighters, communism works and Nancy Pelosi is the epitome of mental stability.

[Wyatt's Torch]

Awesome! One step closer to Skynet.

[MtnMan101]

[Resolute Conservative]

Or Viki.

[wally_bert]

With it being an IBM product, how well would it take me yelling at it when lotus notes freaks out? Notes can’t die soon enough.

BTW, the website Ihatelotusnotes.com is a good laugh.

[mountn man]

Did they use a liberal’s brain as a model?

[Jonty30]

What will be scary about this is that it might eventually figure out how to improve its performance on its own and improve its intelligence.

We could eventually end up with a computer as smart as we are, or smarter, and thinking like a machine.

[Heartlander]

[E. Pluribus Unum]

I wrote a neural network demo in the early '90s that does a pretty good job of demonstrating how a neural network learns to recognize alphanumeric characters.

https://www.box.com/shared/xcnhfu60wc

It was written in QuickBASIC for DOS for an 8088 computer. You can run it in the DOSBox emulator.

http://www.dosbox.com/

[I want the USA back]

“simplified version of the brain’s neural architecture”
note the word “simplified”
Very interesting, since humans don’t understand how the brain works and how the mind works. (brain and mind are two different things)

[captain_dave]

Maybe so, but the brain is not where the mind is. The brain is like computer hardware. Without the software-like Mind (consciousness), it’s just maggot food.