For someone who hates the very idea of a competition, I’m awfully good at them.

Four ML assignments were competitions. I outright won two, with accuracies of 93% (beating the next best by about 1%) and 96% (beating the next best by a fairly wide margin of about 3%). I came in second for the third, with an accuracy of 93.7%, and somehow managed to get dead last in the fourth (I think I screwed up the report, because my cross-validation accuracy was good) with an accuracy of 49%, for which I lost 10 points from the 80 points extra credit I had already accrued 🙂

Update: I’ve been thinking about competitions in general and have come to the conclusion that, sad as they are, they seem to be the only way to call attention to one’s skills before society. My failure to be competitive has only hurt me in things like applications and getting papers published. Therefore, I’m going to begin entering into a few that I believe I have a good chance of winning… perhaps people will start to take notice after I smack a few people’s results into the ground. It’s the risk they take for insisting upon a competition.

Programming competitions are still out, because (a) I already won one eight years ago and (b) I’m being recognized enough as a programmer without requiring additional competitions.