two neuroscience links
First, a discussion of Dostoyevsky's epilepsy. I haven't got much to add; it's interesting stuff, though.
Second, this article on Numenta, the brain-like computing startup from one of the guys behind Palm and Treo. I'd read this Wired article on the startup a little while ago. What they're doing is interesting, and I'm sure they've got some very smart people working for them. But in both articles Hawkins takes pains to poo-poo neural networks as a failed AI technology, and to distance his own "HTMs" from them. Yet after reading the descriptions of his technology (and the training and parameter-tuning it requires) and accounts of the sorts of tasks that HTMs can perform, the tech sounds a lot like multilayer perceptrons. The dog-identifying tasks he points to are really just a more advanced case of the sorts of problems they had us solve in Intro to Neural Networks.
I'm all for moving beginner toolkits out of Matlab and into the open source world — I'm excited about Numenta, and hope it does well. But there's no reason to put down the accomplishments of those who came before you when what you're doing seems to be a variation on the techniques they pioneered.
Comments
I don't think that neural networks are 'failed AI technology', really (although maybe the Numenta people think so, who knows).
As I understand it, the issue is really that a lot of neural-network and perceptron-like learning machines are really just poor or obscure implementations of 'gaussian processes.'
For instance, see here and here; this is a good index site on GPs.
Gaussian processes have their own strengths and weaknesses -- plenty of people use them (and similar 'nonparametric' techniques) today. NNs just got a bit of a bad rap because all the "it's like a brain!" hype sort of obscured the fact that they were special cases of a more general framework that was already well-understood. But once you leave behind all the brain analogies, and understand the math behind them, they're a reasonable technology to use for some approaches/problems.
This also isn't a knock against a lot of 'connectionist' approaches to learning -- plenty of those methods are still popular and in-use today. Numenta's just an example of someone trying to commercialize one of those approaches. It smells a bit fishy to me, but I could easily be wrong...