View Single Post
Old 04-08-2013, 07:26 PM
kafar kafar is offline
Junior Member
Join Date: Apr 2013
Posts: 6
Default Re: *ANSWER* Question 9

Originally Posted by Elroch View Post
I suspect you did not choose your points randomly for the iterative step. If you, for example, always search sequentially through the points for a misclassified point, I believe it is not so uncommon to get caught in a loop of misclassified points. This must eventually resolve itself, but might take a while, and this leads to more long runs and a lot higher average number of iterations. I found that randomly picking misclassified points decreases the average number of iterations a lot and shortens the long tail, presumably by limiting repeated bad choices of misclassified point.
i see it in a different aspect. i won't call them bad choices of misclassified points but less informative. I've been thinking a bit. If the misclassified points are randomly picked, the algorithm has a better chance obtaining "knowledge" contained in the entire sample space, whereas obtaining knowledge by sequentially picking points is always much slower. So slower the convergence.

RE:IsidroHidalgo it was just a guess. i didn't run the experiment to verify. i don't think it would be good idea to look at distance from the start. but after the entire sample has been looked at, **maybe** closer points can be better choices. for the sake of this homework, randomly chosen misclassified points is the way to go.
Reply With Quote