Hoeffding Inequality

[pouramini]

Yes, we don't select h, h is an element of H and we select g

We can't restrict h, then I think emphasizing that h must be fixed before ... is a bit misleading

[henry2015]

Originally Posted by pouramini

Second question:

In "h is fixed before you generate the data set"
I also can't understand your emphasis on "before".

Do you want to say that h shouldn't change?
because I feel h is independent from D then "before" or "after" doesn't mean much. We don't need to have an h in mind to be able to generate D, we can select D, then decide which h to use, then evaluate h over D, but we should use the same h for the test set, right? or maybe h is used somehow in generating D?! Anyway, I think you may mean it should be selected independently from D

I have a similar thought as yours -- the set H is defined/generated independently from D. Hence, defining H before or after choosing D doesn't matter.

However, I think all h in H should be used against D (which is chosen independently from H), and then the learning algorithm will pick one of the h's in H to be g based on the results.

Anyway, I am also just learning this topic; I think we should wait for the "official" comment