In: Advanced Math
Solve SVM for a data set with 3 data instances in 2 dimensions: (1,1,+), (-1,1,-),(0,-1,-). Here the first 2 number are the 2-dimension coordinates. ‘ +’ in 3rd place is positive class. And ‘-‘ in 3rd place is negative class . Your task is to compute alpha’s, w, b. Then, Solve SVM when data are non-separable, using k=2 when minimizing the violations of the mis-classification, i.e., on those slack variables.