Hi Chirag, I read the article out of curiosity and have to say that while they use a few overlapping terms, this is still another regurgitation of Deep Learning networks which bare no resemblance at all to constructs in NuPIC.
The "patter" around it can make it seem as if the two approaches have some degree of alignment - but they don't. For example: > The only method that we knew worked was a brain, so in the long run it had to > be that systems something like that could be made to work,” says Hinton. They then proceed to build something that is totally orthogonal to any appearing in biology - co-opting the approach description of NuPIC without embodying it. The article is interesting for its relaying of the history of perceptrons though. Cheers, David Sent from my iPhone > On Aug 7, 2015, at 6:00 AM, [email protected] wrote: > > Hi All, > Did anyone see this article on tech review about LeCun's New AI framework > at FB. It uses a memory-prediction framework to learn language. That sounds > too similar to NuPIC > > http://www.technologyreview.com/featuredstory/540001/teaching-machines-to-understand-us/ > > > Rgds, > Chirag > > >
