Logistic regression SVM hinge loss

The loss function of SVM is hinge loss: L (y) = max (-T * Y), t = + 1 or-1, which is the label attribute. for linear SVM, y = W * x + B, W is the weight and B is the offset. In actual optimization, W and B are unknown to be optimized, optimize the loss function to minimize the loss function and obtain the optimized W and B.

For logistic regression, the loss function is, because y = 1/(1 + e ^ (-t), L = sum (Y (log (H) + (1-y) log (1-H ))

 

Logistic regression SVM hinge loss