MATLAB Programming Contest Blog

May 9th, 2007

The MATLAB Contest is now underway.

We are in darkness. During the first day of the contest, participants can’t see the scores or the code for any of the entries. This phase challenges participants to create a solution without any feedback from the actual test suite and independently from each other. This phase ends on Thursday at noon EDT and we’ll award our first mid-contest prize for the best entry submitted before twilight.

One other new twist that we didn’t mention in our list of changes is the introduction of complexity into the calculation of the final score. You’ll be penalized if the cyclomatic complexity of your entry rises above 10. Read all about in the rules.

12:02 UTC | Posted in Peg Solitaire | Permalink |

You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

6 Responses to “The Contest is Underway”

1. Alan Chalker replied on May 9th, 2007 at 12:24 UTC :
   

   I’m glad to see the rules changes you’ve made, which should definitely make the contest experience much better. However, I think this begs the question as to how you intend to enforce them. There still isn’t any type of authentication mechanism on the entry submission page, which means that somebody who wants to still welch the contest or spam-tweak or probe out the test suite could do it anonomously.

2. Matthew replied on May 9th, 2007 at 12:46 UTC :
   

   Hi Alan and welcome back.

   Your comment about enforcement is mostly true. We have a couple of steps we can take if someone is being difficult, but they’re fairly limited.

   We’re mostly counting on our participants to cooperate. In the past, people submitting entries like these have been white hat hackers like yourself. It was all in good fun and we haven’t seen evidence of real malice. Hopefully this spirit will continue in this contest under the new rules.

   If anyone notices something shady going on, please let us know right away at contest@mathworks.com.

3. Heinrich Acker replied on May 9th, 2007 at 14:05 UTC :
   

   There are a couple of nice new ideas in this contest! I really like the problem. What you have written aubout group intelligence and the white flag is clever, no real gentleman can welch you after this. The complexity-depended score is also interesting, although it should always be possible to avoid a penalty by using lots of functions. Let’s see what happens.

   Last not least: The GUI provided with the testsuite is the best so far. It will allow spectators like me to follow the evolution of solutions in this contest.

4. Alan Chalker replied on May 9th, 2007 at 15:11 UTC :
   

   I’m posting this both here and in the newsgroup to ensure the contest team sees it:

   There is an error in the grade function, but I haven’t been able to figure it out yet.

   Try to run the examples in the rules like this:

   grade([7 0 2 0; 0 0 8 0; 0 7 0 0; 0 0 3 4], [1,3,3,3])
   result = 31
   according to the rules it should be 25

   likewise

   grade([7 0 2 0; 0 0 8 0; 0 7 0 0; 0 0 3 4],[4,4,4,2;4,2,2,2;2,2,2,4]);
   result = 35
   according to the rules it should be 17

   I only discovered this because as I was testing things that should work out, my score was going up, not down.

5. Matthew replied on May 9th, 2007 at 16:43 UTC :
   

   Thanks Alan. We’ve posted an update and announced it on the newsgroup.

6. Alan Chalker replied on May 9th, 2007 at 17:21 UTC :
   

   I just tried the updated grade.m and it doesn’t seem to be fixed. You did this:
   Old:
   f = moves(i,[1 2]);
   t = moves(i,[3 4]);

   New:
   t = moves(i,[3 4]);
   f = moves(i,[1 2]);

   The correct New (based upon the newsgroup discussion) should be:
   f = moves(i,[2 1]);
   t = moves(i,[4 3]);

   Alternatively, you should update the rules to indicate the reversed order.

Leave a Reply

<pre class="code">
a = magic(3);
sum(a)
</pre>

These postings are the author's and don't necessarily represent the opinions of The MathWorks.