Well, I managed to get the proving problems sorted out (tech support for the uni end and proper username on the RL Comp end).
Here are the results:
Total Number of Episodes: 24392
Total Return: 4782563.0
Total Time taken: 24 hours, 36 minutes, 2 seconds
These results mean nothing unless put into context, i.e. my position on the leaderboard. As at 19th June 2008, I am 4th. The top 5 are as follows:
1 The Cobras 6474680.0 Mon Jun 16 19:10:40 -0600 2008
2 Neural Information Processing Group, Eotvos Lorand University 6096040.0 Wed Jun 04 05:32:14 -0600 2008
3 Loria INRIA – MAIA 5366230.0 Wed Jun 11 17:00:04 -0600 2008
4 SmartCraft 4782560.0 Wed Jun 18 17:52:11 -0600 2008
5 UIUC CS548 4591890.0 Tue May 13 08:31:51 -0600 2008
Good results, but I am 4th. This means I need to focus my efforts into the project once more and make it better. The gap from 4th to first is quite large too, and I hope it can be done. Time to trial out probabilistic play and also look at the problem of bell-shaped fields (a somewhat common problem).
I have 5 or so days to do this, so I need to work fast and without distraction. But, other events currently take my time (Taekido), so I’ll have to start tomorrow. Doesn’t mean my mind stops though.