Jump to content
  • Advertisement

Archived

This topic is now archived and is closed to further replies.

fup

Stephen Wolfram book

This topic is 5968 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Not AI as such but this may have some major implications. It looks like it''s going to be a fascinating read anyway. Mine''s on order. I thought some of you guys (and gals?) might appreciate this link. http://www.wired.com/wired/archive/10.06/wolfram.html?pg=1&topic=&topic_set=

Stimulate

Share this post


Link to post
Share on other sites
Advertisement
Yep, a hearty tome, indeed.

I have no doubt that cellular automata are a great and wonderous thing (I personally believe that CAs could solve the protein folding problem - Timkin, I''d love to hear your thoughts on that one.), however, Wolfram has sacrificed modesty for the sake of advancement.

This book represents 15 years of work done by Wolfram himself, however, it is almost entirely unpublished work. Unpublished = not peer reviewed. (OK, I''ll admit that Calculus was not accepted until some 100 years after Newton''s death.) In addition, it is published by Wolfram Media. Again, no peer review.

I''m also a little troubled by the tone of the introduction. Wolfram seems to be reaching out and saying, "Look! Look what I''ve done!"

Regardless, it should be a hearty read and I''m sure some great steps forward will be made as a result. Am I willing to throw away all other scientific knowledge to take up the banner of conversion? Hmmmm....maybe......


-Kirk

Share this post


Link to post
Share on other sites
And what have he done with CAs?

I don''t know too much about them (just what I''ve studied in my Automatas subject) but if it is so important...

Share this post


Link to post
Share on other sites
I rather like the idea he''s doing it without peer review. Nice to see someone with so much confidence (regardless of how correct his ideas are).
Anyway, he''s certainly going to meet with a lot of resistance from the academics. Especially the ones who secretly envy Wolframs undoubted talent (and the way he''s turned it to considerable profit).

Here''s another link to an interesting review my Kurzweil

http://www.kurzweilai.net/meme/frame.html?main=/articles/art0464.html





Stimulate

Share this post


Link to post
Share on other sites
fup,

You''ve got a good point there with the peer review issue.

I agree in the idea the CAs are very interesting in a huge number of areas. Take a look at Craig Reynold''s site (http://www.red3d.com/cwr/) and the demos there. The whole Boids phenomenon is CA. There''s also some great stuff on CA attractors which suggests application to Pattern Recognition/Classification. Ahhh....nice stuff!!


-Kirk


Share this post


Link to post
Share on other sites
My book''s on order too, but I''ll explain what I''ve been able to glean from the various articles I''ve read:

Wolfram studied the 256 rules possible for a 1-D cellular automata (central bit and surrounding bits correspond to 256 permutations for the new corresponding bits), and found that the 110th rule created some interesting results. Then another CA researcher found that this rule was in fact a Turing-complete method of calculation (I forget what the criteria for this is, but it shouldn''t be too hard to find). From this rule, he''s spawned all sorts of different manners of computing things, and now he''s suggesting that CA''s can represent everything in the universe, "in only four lines of code in Mathematica."

I don''t know if within those 1200 pages there''s something actually linking these two ideas (or if it follows logically), but that''s what I''m ordering the book for

When you all get a chance to look at a copy of it (especially people with more of a background and/or knowledge of CA than myself), post your thoughts on it.

Share this post


Link to post
Share on other sites
Here''s another link to an interesting CA (among other things) site.

http://www.santafe.edu/~wuensch/ddlab.html


-Kirk

Share this post


Link to post
Share on other sites
I think I agree with Kurzweil, there''s only so much complexity there.

And besides, what if Wolfram is right? Long term emergence is not really much use to AI developers -- let alone game developers, though there may be some fun Ph.D''s to be done there.

Alex



Artificial Intelligence Depot - Maybe it''s not all about graphics...

Share this post


Link to post
Share on other sites
I usually find that I get new ideas when I read books which discuss loosely related (or even unrelated) subjects. I wouldn''t have had a lot of the ideas I have had if I just stuck to AI literature.

That''s why I consider them useful...




Stimulate

Share this post


Link to post
Share on other sites

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!