Im not sure if this theory goes all the way to AGI, but I cant think of any
better way to attempt it yet, it definitely would make a very smart artificial
intelligence tho definitely, would put a tidal wave in a little fish pond (like
where we currently sit).
But I cant imagine myself past the smaller machine metric (which i call okhams
metric) to make the computer not just spit the same data back out at you,
offering no more.
Whats different about implementing this than an ordinary evolving network
(otherwise where is the evidential success in our environment!?!?!?) is I think
most computer scientists opt for backprop in these cases, and possibly you dont
get the amazing results if you do it a harder way, but I remember hearing Jeff
Clune talk about "pick most simplest model" But maybe a little sportily tell
you perhaps it has been performed poorly to date! So its still on the books
for a look, just too much work on the plate, mayte!!!
------------------------------------------
Artificial General Intelligence List: AGI
Permalink:
https://agi.topicbox.com/groups/agi/T36c83eb0aa31fc55-M81ed676a0147b1d9927f41e3
Delivery options: https://agi.topicbox.com/groups/agi/subscription
