On Sun, Dec 18, 2005 at 09:55:32PM +0100, Shane Legg wrote: > To my mind the key thing with neural networks is that they > are based on large numbers of relatively simple units that > interact in a local way by sending fairly simple messages.
That's just one facet. These are units of a large number of classes, with different message types and *changing connectivity*, depending on internal and external (rich) state (morphogenetic code). > What you seem to be criticising in your memo is what I'd call > "feed forward neural networks". Of course Minsky & Papert in 1969 have shown that all ANNs are useless, by virtue that single-layer perceptrons can't solve problems that are linearly inseparable. The inescapable logic of it has killed ANN R&D funding for about 15 years. -- Eugen* Leitl <a href="http://leitl.org";>leitl</a> http://leitl.org ______________________________________________________________ ICBM: 48.07100, 11.36820 http://www.ativel.com 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE ------- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
signature.asc
Description: Digital signature
