13-PracticalMachineLearning

Clopinetcomisabelleprojectssvmapplisthtml some people

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: separa%ng hyper plane x1 Perceptron •  •  •  •  x: data point y: label w: weight vector b: bias +1 - 1 w x1 x2 w1 w2 w3 x3 b - 1 The XOR Problem x3 x2 x1 Support Vector Machine •  Widely used for all sorts of classifica%on problems www.clopinet.com/isabelle/Projects/SVM/applist.html •  Some people say it is the best of the shelf classifier out there Maximum Margin Classifica%on x2 x2 x1 x1 What about outliers? : slack variables x2 x1 XOR problem revised x=0 x=0 Did we add informa%on to make the problem seperable? Polynomial Kernel in 3D Quadra%c Kernel Kernel Func%ons •  Polynomial: •  Radial basis func%on (RBF): Kernel Trick for SVMs •  Arbitrary many dimensions •  Li4le computa%onal cost •  Maximal margin helps with curse of dimensionality SVM Applet h4p://www.ml.inf.ethz.ch/educa%on/ lectures_and_seminars/annex_estat/...
View Full Document

Ask a homework question - tutors are online