Monday, November 22, 2010


Thanks to everyone for the useful feedback on my Laser game so far. I have added another three levels, making a total of ten. I am trying to order the levels in increasing difficulty. However, I am having trouble comparing the difficulty of levels (since I already know the solution to levels when I design them, it is hard to judge how hard it would be for someone else to solve them). It would be great if I could get some more feedback on if you encounter an earlier level which appears to be more difficult than a later level.

Thursday, November 18, 2010




My laser reflection game now has seven levels. I have found that it is much easier to create levels for this game than it was for the art gallery problem game. I think the last two levels are quite challenging, so let me know if you manage to beat them.

Sunday, November 14, 2010


Today I started working on a new Java game. It will be a geometry-based puzzle game. The idea is to position mirrors so that they reflect lasers to a set of targets. I have posted a prototype version here.

State of the Art

For my master's thesis I have been looking into compression-based techniques for classification, clustering, and anomaly detection. I have implemented a classification algorithm using PAQ and evaluated it on three datasets. So far the results look very promising and it seems to get state of the art results on all three datasets. The first dataset (called 20news) involves categorizing newsgroup articles into one of twenty categories. The other two datasets are spam-filtering (ling-spam and PU1). For the spam-filtering datasets a tradeoff can be made between spam/ham misclassification rates. I therefore evaluated my algorithm using ROC curves. Although my algorithm wasn't the best spam-classifier on all portions of the ROC curves, on both datasets there was still a significant portion in which it got state of the art results.