On Fri, 10 Sep 1999, J. R. Molloy wrote:
> Robert J. Bradbury wrote,
> <>Interestingly enough the article comments on the fact that de Garis
> >has his 72 gate array [= 70 million brain cell or 10,000 500 MHz
> >Pentiums] built but they are having problems figuring out how to
> >connect together the 60,000 circuitry blocks to actually "do something".
> Take at look at self-assembling wires from HP: + long discussion.
I guess taking the statements out of context loses something. John, if your response was indended for the "how to connect" part of the statement, then I've misexpressed myself.
The problem isn't "wiring them together" but having them "do something" once wired. Nobody has ever had a 70 million cell neural net before. How deep do you make it? How broad do you make it? How many inputs & outputs should each neuron have?
I'm not sure how flexible the de Garis architecture really is but if the depth can be up to 10 deep, and the net has 1000 inputs and 100 outputs, and each neuron can interact with 500 others, the possible number of configurations is huge. [I'm not an expert on neural nets, so if someone can explain this better please do so.]