View Single Post
  #1  
Old 07-10-2014, 02:53 AM
rabbitdoll rabbitdoll is offline
Junior Member
 
Join Date: Jul 2014
Posts: 1
Default Out of sample error for linear noisy target

I am confused about how to derive the Eout=(1+(d+1)/N)\sigma^2 for the learning curve of linear target error, as shown as the third equation in http://work.caltech.edu/library/084.pdf ?

Could anybody help me?

Thanks!
Reply With Quote