Thread: problem 1.7
View Single Post
Old 09-15-2016, 06:54 AM
iseedoug iseedoug is offline
Junior Member
Join Date: Sep 2016
Posts: 1
Default Re: problem 1.7

While this is important, I think it is easier to not get stuck on it being "training error". You can simply view the binomial distribution with getting 0 heads on 10 flips for 1 coin. This will also give you a training error of 0 in this case. (because nu = 0).
Reply With Quote