Question: Consider exact kernel ridge regression with polynomial kernel of the form: K ( x , y ) = ( x y + c )? (2)

Consider exact kernel ridge regression with polynomial kernel of the form: K ( x , y ) = ( x y + c )? (2) for some constant c > 0. Assume that the size of the d-dimensional training set N satisfies: N >> (). Propose exact kernel ridge regression method for training with time complexity o( N3). Explain what is time complexity of your algorithm. Hint: You can use the following Sherman-Morrison-Woodbury identity on matrices: (B+aUV )-=B- -aB U(I+QV B 'U)- 'VB-1, (3) where a is a scalar
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
