I could not understand what is the purpose of the generalization error bound, why do we need to calculate it?!. How does the generalization error bound work with 1-nearest neighbour algorithm ?. Does this mean, there is a guarantee that the error rate will be less than 1-nn on test set. Please see the image for the generalization error bound that I could not understand it. enter image description here

This is the paper: https://pdfs.semanticscholar.org/a08f/133198a845589e6801e2125dc973d2652d66.pdf

  • $\begingroup$ @Dougal do you know about it ?! $\endgroup$ – miss Ran Dec 2 '19 at 11:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.