Tweet Notes (CSE 494/598 F11)
This will contain collectively generated notes
Monday, December 5, 2011
In Parametric learning the size of the learn representation is separately described and is not promotional to the training data.
E.x: Linear Classifiers.
Non parametric learners keep entire training data around.
Problem with K-NN is that it doesn't do well in high-dimensions.
Share to Twitter
Share to Facebook
Share to Pinterest