On 23 Aug 2012, at 18:11, benjayk wrote:
Or how can you determine whether to program a particular program or
not? To
do this computationally you would need another program, but how do you
determine if this is the correct one?
You don't.
In theoretical inductive inference theory (Putnam, Gold, Blum, Case
and Smith, Oherson, Stob, ...), you can show this:
The more machines can be wrong, and relies on other machines, and
change their minds, and (amazingly enough) change their mind wrongly
(changing a correct theory for the know facts by an incorrect one),
from times to times, the more larger are the classes of phenomena that
they are able to recognize (build correct theories fro them).
And those theorem are non constructive, meaning that in the world of
inference inductive machine, a machine capable of being wrong is
already non computably more powerful than an error prone machine.
if like some people, you define intelligence by an ability to
extrapolate series, there is something transcendentally (non-
computably, even with strong oracles) more powerful than any inference
inductive machine, it is a couple of inference inductive machines.
All machines evolves, in a non predictable way, through errors, notably.
Bruno
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.