Design Con 2015
Breaking News
News & Analysis

IBM demos cognitive computer chips

8/18/2011 04:01 AM EDT
28 comments
NO RATINGS
Page 1 / 2 Next >
More Related Links
View Comments: Newest First | Oldest First | Threaded View
<<   <   Page 2 / 3   >   >>
markhahn0
User Rank
Rookie
re: IBM demos cognitive computer chips
markhahn0   8/18/2011 11:39:22 PM
NO RATINGS
I'm disappointed eetimes doesn't have something more technical about the architecture here. are the weights stored in registers? the device is said to be all digital, so doesn't an update cycle require a lot of flops? isn't a fan-out of 10,000 pretty hard to drive? why is 10Hz an adequate update rate (it certainly doesn't match real neurons.) if this thing is structured like an xbar, does that mean there's some kind of multiplier at each crossing? (the weights and summation are digital, right?)

old account Frank Eory
User Rank
Rookie
re: IBM demos cognitive computer chips
old account Frank Eory   8/18/2011 9:44:36 PM
NO RATINGS
A pure digital chip that is supposed to emulate neurons, synapses and dendrites? As mdkosloski said, they might as well just do the whole thing in software. Oh wait, they did. So I guess the point of building the chip was just so they could eventually build a 10B neuron/100T synapse machine the size of a shoe box? They could simulate that in software too...it just takes more computers to do it.

R G.Neale
User Rank
Rookie
re: IBM demos cognitive computer chips
R G.Neale   8/18/2011 9:39:37 PM
NO RATINGS
Mr Rbtrob- There were two parts to the Stanford synapse experiment. The first was to demonstrate 100 level resolution. The second to demonstrate that the synapse characteristics could be reproduced, for this they used 15 pulse trains. I reduced it to fewer for the purpose of my explanation and illustration. I think the use of epitaxial regrowth of the same crystals, as illustrated in one of my figures is the way to obtain reproducibility and 100 level resolution. However, that path tend to lead to the conclusion that PCRAM might offer a superior solution.

BrainiacV
User Rank
Rookie
re: IBM demos cognitive computer chips
BrainiacV   8/18/2011 9:15:16 PM
NO RATINGS
I dunno, it's hard to tell what you've taught a neural circuit. It's easy teaching it the difference between a bruised and unbruised apple. I once heard the story of how they tried to teach a neural program to detect tanks on a field. They showed it fields with and without. When it came for a field trial it failed spectacularly. Going back over the input data, the best guess is that they had taught it how to tell a sunny day from a cloudy day, which turned out to be the difference when the tanks were on the field or not.

DataBass
User Rank
Rookie
re: IBM demos cognitive computer chips
DataBass   8/18/2011 9:07:14 PM
NO RATINGS
Will the device be made out of a platinum/iridium alloy?

Code Monkey
User Rank
Rookie
re: IBM demos cognitive computer chips
Code Monkey   8/18/2011 8:58:41 PM
NO RATINGS
To the lab guys: If you drop the chip and it breaks, don't substitute the one marked "Abnormal - Do Not Use".

pixies
User Rank
Rookie
re: IBM demos cognitive computer chips
pixies   8/18/2011 8:55:44 PM
NO RATINGS
$20M for Phase 0 and Phase I combined, not bad. I wonder what's the goal for Phase II. It will be exciting to see a machine with 10B neurons built and find out what it can do.

rbtbob
User Rank
Rookie
re: IBM demos cognitive computer chips
rbtbob   8/18/2011 5:43:05 PM
NO RATINGS
On reading your explanation of what Stanford had demonstrated, I was amazed that they could perform a resistance change using such a large number of pulses AND, if I understand your analysis, get the device to repeat the cycle enough to demonstrate a workable functionality. Now the question is whether IBM is pursuing Stanford's scheme or some other? Also, could directional current pulses be used to enable both additive and subtractive resistance changes?

R_Colin_Johnson
User Rank
Blogger
re: IBM demos cognitive computer chips
R_Colin_Johnson   8/18/2011 5:23:33 PM
NO RATINGS
This is IBM's first generation device, intentionally created to transfer its supercomputer simulations to a hardware platform. As their simulations become more detailed, IBM will have to deal with all the mentioned issues going forward. (And yes, it is relatively easy to program a computer to play Pong or recognize a numeral, which is why these were good metrics for a very simple chip learning a task on its own.)

Volatile Memory
User Rank
Rookie
re: IBM demos cognitive computer chips
Volatile Memory   8/18/2011 3:45:22 PM
NO RATINGS
rbtbob: Axons and synapses work. PCM doesn't. As demonstrated by JPL/NASA, Samsung, and even Numonyx/Micron.

<<   <   Page 2 / 3   >   >>
Flash Poll
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Top Comments of the Week