In article <317EDD47.446B9B3D at etl.go.jp>, Paultje Bakker <bakker at etl.go.jp> writes:
|> > One last point - to build a nonhuman intelligence, build a huge
|> > cortex and connect it to some inputs. The difficulty will come in
|> > training it - deciding which higher level concepts it builds are
|> > important.
|>|> Now why didn't I think of that!
Cos you didn't work out what cortex does, exactly.
|> If I were you, I would apply for a patent, and quick. You
|> don't know who might be reading these posts.
I think there has to be something a bit more specific to get a patent. I don't
know what the learning algorithm is, even though I have some good constraints on
what it must look like. Anyway, I'll be happy if someone proves my theory correct
by building a machine that implements my ideas.