MATLAB Programming Contest Blog

November 7th, 2008

Doug has posted his commentary on the first King of the Hill.  It is in video form on his regular Video tutorial blog.  The first four boards are run and commented on.  (Six minutes).

We can’t get the Queue and Top 20 page to refresh. For now, please use this (ugly) page here.
–Matthew

22:06 UTC | Posted in Army Ants, MATLAB Contest | 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 “Contest video commentary: Darkness”

1. the cyclist replied on November 8th, 2008 at 03:36 UTC :
   

   Thanks to Matthew for working some serious overtime trying to keep this contest going. I don’t know about anyone else, but the auxiliary web site is not working properly for me. Strangely, it sometimes took me to what seemed to be a current queue, but most of the time is takes me to the queue as of 8:12 PM on 11/7. Weird. Also, I can no longer respond to the newsgroup thread. It takes me to an error page.

   Also, thanks to Doug for a great video presentation on the king of the hill.

2. Alan Chalker replied on November 8th, 2008 at 05:05 UTC :
   

   I’m seeing the same issues, however the queue is still operational. You can still submit and see results in the chronlogical listing - albeit after a delay.

   I have to say I’m very impressed that the run times of the leading entries have dropped so significantly. We’re now at the 60s mark, which surprises me a lot. I guess I need to dig around in the profiler a bit more to understand why this is.

3. Alan Chalker replied on November 8th, 2008 at 05:10 UTC :
   

   Is Nathan q’s winning entry from twilight the King of the Hill now or is Jan Langer’s darkness winner? The queue page says it’s still Jan’s

4. Seth Popinchalk replied on November 8th, 2008 at 13:20 UTC :
   

   The queue page indicates that nathan q is now the king of the hill.

5. Alan Chalker replied on November 8th, 2008 at 14:56 UTC :
   

   For those of you interested in understanding how the leading solvers are working, the current top entry (Daylight Again 31) is a heavily commented version I created that should explain things pretty simply: http://www.mathworks.com/contest/armyants.cgi/view_submission.html?id=50436

   It’s surprisingly short and simple to understand. Hopefully this will make it less intimidating for those of you wanted to jump in and compete over the weekend.

   Credit goes to Nick Howe I believe who originally submitted this basic algorithm in his Cowardly Lion entry, which was subsequently improved upon by SY and David Jones before I took a stab at it.

6. Alan Chalker replied on November 8th, 2008 at 16:31 UTC :
   

   Matthew:

   Since we can’t post to the newsgroup, I need to reply to your comment there here. Looking at the way the submissions are going, time is less of a concern now than it was before, although I think it’s still something to try to find a solution to because it’s forced us into a ‘tweak war’. Interestingly enough, the contest system is very sensitive to the random inputs. Very slight changes in the random thresholds are producing very large reductions in the results, unlike anything I’ve ever seen in previous contests. Unfortunately that probably meaning we’ll degenerate into a queue spamming at some point. I’m not sure there is anything that can be done about this, although I know if previous contests there have been a lot of suggestions on how to minimize the effect of the random variable seeds.

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.