2017-10-12
HW2 sk_test doubt.
Prof,
On slide 11 (Sep 27), it is mentioned that our SVM will compute:
`g(vec{x}) = sign(sum_{i in I} alpha_i y_i K(vec{x}, vec{x}'_i) + (B - A)/2)`
and y is defined as
`y_i = 1` if `x_i in X^+` and `y_i = 0` otherwise.
But with this definition,
won't our SVM simply ignore negative support vectors?
(Edited: 2017-10-12) Prof,
On slide 11 (Sep 27), it is mentioned that our SVM will compute:
@BT@g(vec{x}) = sign(sum_{i in I} alpha_i y_i K(vec{x}, vec{x}'_i) + (B - A)/2)@BT@
and y is defined as
@BT@y_i = 1@BT@ if @BT@x_i in X^+@BT@ and @BT@y_i = 0@BT@ otherwise.
But with this definition,
won't our SVM simply ignore negative support vectors?
It should be `y_i = -1` otherwise. I have fixed it now in the slides. Sorry about that.
It should be @BT@y_i = -1@BT@ otherwise. I have fixed it now in the slides. Sorry about that.
(c) 2024 Yioop - PHP Search Engine