News for hackers. Stuff that matters.
Re:Yann LeCun on the IBM neural net chip by itg on Saturday August 09, 2014 @02:58PM (#8157703)
>My main criticism is that TrueNorth implements networks of integrate-and-fire spiking neurons...Spiking neurons have binary outputs (like neurons in the brain).

Isn't this better. LeCun is looking at this only from a machine learning perspective.

Re:Yann LeCun on the IBM neural net chip by modeless on Saturday August 09, 2014 @03:59PM (#8157867)
Machine learning is the goal. This chip should be judged on its learning performance, not on how well it adheres to an oversimplified and incomplete model of how biological neurons might work in the brain.
Re:Yann LeCun on the IBM neural net chip by oxtopus on Saturday August 09, 2014 @05:40PM (#8158083)
FWIW, there is no learning on-chip. Machine learning is not the goal of this project, nor is its success dependent on it's learning capabilities (at least not at this phase). Where it does succeed, however, is in low-power computation in an architecture that is scalable and fault tolerant. LeCunn is criticizing an orange for not tasting like an apple.
Re:Yann LeCun on the IBM neural net chip by modeless on Saturday August 09, 2014 @05:56PM (#8158124)
Learning may not happen on-chip, but the network is still learned, and the performance of the chip is dependent on the learning. The spiking architecture of the chip means that the best learning algorithms can't be used. An ASIC implementing a convolutional neural net could also be low-power, scalable, and fault-tolerant, while taking advantage of the best currently known learning algorithms and ultimately performing a lot better on real tasks.
Re:Yann LeCun on the IBM neural net chip by gamegoblin on Saturday August 09, 2014 @03:28PM (#8157799)
Just because the brain does it one way, doesn't mean it's the best way. Birds versus jets, nature versus artificial.

That being said, the brain is basically the only working example of true intelligence we have, so perhaps trying to emulate it isn't a bad idea.

But I have this sci-fi notion that eventually the AI community will produce some sort of intelligence that is unimaginably different from our current notion of a brain.

Re:Yann LeCun on the IBM neural net chip by Estragon on Saturday August 09, 2014 @03:58PM (#8157865)
The neurons in the IBM chip are only slightly better approximations to biological behavior, nowhere near close enough to hope that it will accidentally result in brain-like intelligence.

Last I heard, the behavior of biological neurons was so badly understood that even the behavior of the 300-neuron C. elegans worm could not be accurately simulated, even though the neuron connections have been fully determined. That was about 8 years ago, though.

Re:Yann LeCun on the IBM neural net chip by sanxiyn on Saturday August 09, 2014 @04:53PM (#8157986)
C. elegans situation remains the same today, although there now is a serious project to get it done.

http://nemaload.davidad.org/

Re:Yann LeCun on the IBM neural net chip by Estragon on Sunday August 10, 2014 @09:06AM (#8159631)
Thanks, this is the coolest thing I've seen in weeks.
Re:Yann LeCun on the IBM neural net chip by p1esk on Wednesday August 13, 2014 @05:30PM (#8175113)
I believe that project is dead. David Dalrymple took a job at Twitter a few months ago.
Re:Yann LeCun on the IBM neural net chip by DaniFong on Saturday August 16, 2014 @09:30PM (#8188260)
Am close to the source; the project is not dead, despite his full time employment, it has made some recent progress and is currently seeking funding.
Re:Yann LeCun on the IBM neural net chip by nabla9 on Saturday August 09, 2014 @03:20PM (#8157776)
When I heard about spiking neurons, I assumed that IBM tries to speed up Blue Brain project or cognitive research in general.

But it seems like SyNAPTIC group tries to push this chip into common machine learning applications.

http://www.research.ibm.com/articles/brain-chip.shtml

Re:Yann LeCun on the IBM neural net chip by sanxiyn on Saturday August 09, 2014 @03:46PM (#8157843)
I should note that Blue Brain Project is very critical against the group which produced TrueNorth. The director of Blue Brain Project wrote an open letter titled "IBM's claim is a hoax", and considers the approach completely useless for neuroscience research.

http://technology-report.com/2009/11/neuroscience-expert-dr-...