Registered: Oct 2004
Posts: 3

In NKS, there is much use of analysis and ranking of programs (P, 1105). Are there publicly available applications or sets of code to rank competing programs on the basis of user-set criteria?



cm

Report this post to a moderator | IP: Logged

Registered: Oct 2003
Posts: 46

The picture on page 1105 http://www.wolframscience.com/nksonline/page-1105 where all 256 elementary cellular automata are played in a game against one another are produced using a few lines of Mathematica code.

I don't know of any other programs to do this, but of course the idea is pretty new.

Report this post to a moderator | IP: Logged

Registered: Oct 2004
Posts: 3

Richard:

Thanks for the reply. I don't think that setting up a competition is all that tough, but I was hoping that some people who have invested some time into the topic may have some highly efficient approaches, especially if the number of rules were very large.

I'm looking at testing a large number of rules and would prefer to do it in an optimal way.

cm

Report this post to a moderator | IP: Logged

Registered: Oct 2003
Posts: 46

Oh, I see your point.

I guess there might be. In general I suppose it could be difficult, since you're basically just doing a one-to-one comparison. It's not like there is some hierachical structure to the problem where you know you can prune things.

There may well be some clever tricks though. Perhaps looking at the forms of the programs in particular cases would help?

Report this post to a moderator | IP: Logged