Ben,
A further thought occurs to me about how learning derives from practice - &
I wonder whether you've thought about this. Basically, in humans and
animals, the kinds of learning we've been talking about - learning to
crawl, manipulate, walk, talk, play tennis forehands or pianos - involve not
just practice but development. The body and brain have not only to discover
appropriate trajectories of movement, (and associations of ideas), but they
have to wire themselves anew, and grow muscles in order to support those
trajectories and associations.The body and brain have to develop/evolve. And
this is considerably more complex than a computer making new memory
connections.
(My impression is that development is usually left out of scientific
approaches to learning - it's so complicated).
But if organisms learn in this machine-morphing way, then there must be
considerable advantages to it - & it may even be necessary for AGI. Have you
thought about this area?
Ben:It seems to take tots a damn lot of trials to learn basic skills, and we
have
plenty of inductive bias in our evolutionary wiring...
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=8660244&id_secret=71096801-5b1a04
