For the simplest kernel classification rule, what choices of the smoothing parameter h are analogous to selecting
Question:
For the simplest kernel classification rule, what choices of the smoothing parameter h are analogous to selecting kn = 1 and kn = n, respectively, in the nearest neighbor rule? What happens to the error in these extreme cases?
Fantastic news! We've Found the answer you've been seeking!
Step by Step Answer:
Related Book For
An Elementary Introduction To Statistical Learning Theory
ISBN: 9780470641835
1st Edition
Authors: Sanjeev Kulkarni
Question Posted: