CSC 2515 2008
Lecture 10
Support Vector Machines
Getting good generalization on big datasets
Preprocessing the input vectors
A weird measure of model complexity
Preventing overfitting when using big sets of features
What to do if there is no separating plane
A picture of the best plane with a slack variable
Why do large margin separators have lower VC dimension?
A potential problem and a magic solution
What the kernel trick achieves
Support Vector Machines are Perceptrons!
Learning to extract the orientation of a face patch (Ruslan Salakhutdinov)
The root mean squared error in the orientation when combining GP’s with deep belief nets