Thread: Q3 linear model
View Single Post
  #2  
Old 05-04-2013, 10:44 PM
yaser's Avatar
yaser yaser is offline
Caltech
 
Join Date: Aug 2009
Location: Pasadena, California, USA
Posts: 1,477
Default Re: Q3 linear model

Quote:
Originally Posted by Michael Reach View Post
I am having some trouble understanding this question, and making sure it isn't a trick. I note that in lecture 8, the professor gave an equivalent of VC dimension for linear regression - but in quotes, saying that the regression example doesn't really have a VC dimension like classifications do. Did I misunderstand? Is this question asking for the VC dimension or the "VC dimension" equivalent? Is the right answer something like d+1, or should I be choosing 0 or infinity or something?
Questions 2 and 3 are about linear classification, so the VC dimension in this case is genuine. Thank you for bringing this up, as some people may be confused after seeing Question 1 about linear regression and not noticing "decision boundary" in Question 2.
__________________
Where everyone thinks alike, no one thinks very much
Reply With Quote