On 9/6/06, Pei Wang <[EMAIL PROTECTED]> wrote:
> > Hawkin's theory is essentially a variation of MLP (multi-layer perceptron).
> > The problem of variables does not go away. Once you try implementing NN on
> > vision you'll see why...
> >
> > YKY
>
> Though Hawkin's book is vague in technical details, he makes it clear
> that he disagree with NN on fundamental issues.
>
> His recent work with Dileep George on Hierarchical Temporal Memory is
> much closer to Bayesian network than to MLP or any NN.
> > The problem of variables does not go away. Once you try implementing NN on
> > vision you'll see why...
> >
> > YKY
>
> Though Hawkin's book is vague in technical details, he makes it clear
> that he disagree with NN on fundamental issues.
>
> His recent work with Dileep George on Hierarchical Temporal Memory is
> much closer to Bayesian network than to MLP or any NN.
Well, let's change the question a bit: are there some specific features in Hawkin's theory that you (or others) think should be included in my architecture to make it better?
In general, I'd like to hear why people disagree with my architecture; and if not, we can explore some options for collaborating. That would be more interesting (to me at least) than talking about a third person's theory without a focus.
YKY
To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
