Originally posted by holzpusherI was just wondering how would you go about programming AI.
Yes I know about programming in Python. I've been doing it for a couple of years but I'm not yet a genuine guru. What do you want to know?
Are there any good websites I can go to to teach my self?
Thanks
Oh, and if you need to learn the concepts of AI-programming I would
recommend the book: "Artificial Intelligence - A modern approach", by
Stuart Russel and Peter Norvig. Depending on your previous level
education it may be a tough read, but it's definitely worth it. Best book I
own on the subject.
Though I really liked "Introduction to Expert Systems", by Peter Jackson
as well.
These books are more about the general concepts and ideas rather than
actual programming (though Expert Systems cover some Lisp
programming). There's lots of mathematics involved and pseudo
code examples, but you're supposed to take it to your language of
choice (Python in your case).
Here's an extract from the Introduction of "Artificial Intelligence - A
modern approach" that I really like (it kept me reading even when it got
very abstract and complicated):
"AI addresses one of the ultimate puzzles. How is it possible for a
slow, tiny brain, whether biological or electronic, to perceive, understand,
predict, and manipulate a world far larger and more complicated than
itself? How do we go about making something with those properties?
These are hard questions, but unlike the search for faster-than-light
travel or an antigravity device, the researcher in AI has solid evidence
that the quest is possible. All the researcher has to do is look in the
mirror to see an example of an intelligent system."
Granted, when I look in the mirror the last assertion doesn't quite work,
but for most people I suspect it does. That's what's so fascinating about
AI (in my opinion). That, indeed, it should be possible with enough
computer power and correct algorithms. And, as you'll learn if you really
are interested in this, the algorithms are already here. Faster computers
(with enough memory to store computational data) is the only thing
missing to create a truly self-reliant, intelligent system.
Expert systems are already in use successfully in various parts of
medicine (and chess, as I'm sure you know). An expert system is limited
to a very specific area of expertise though. You can't really have a
conversation about the weather with such a system. (In fact, I think
you'd be surprised if you knew exactly how much computing power is
required just to talk somewhat intelligible about the weather and other
things we humans consider simple and trivial in nature.) A Turing
complete AI system would be such that you can't tell whether you're
talking to an actual human or a computerised system in a chat room, for
instance (though when chatting you sometimes get so stupid replies that
it would be hard to tell even a tin can from an actual human).
Originally posted by eamon oSpammer! 😛
this was a good one. incidentally, foreigners, spam was a nasty tinned meat in post war days.
i remember this sketch when it was first on tv. still good.
most people never missed an episode of the weekly madness.
http://www.youtube.com/watch?v=ODshB09FQ8w
the little red riding hood sketch also on youtube is something else 🙄