Please solve so that I can learn.

Problem 3 (30 points) 1. Find the VC-dimension of the function class H1 = {ha; a ER}, where ha : R - {0, 1} is defined as 1, if x E a, a + 1] ha(2) = 1, if x E [a + 5, a + 6] 0 otherwise.2. Find the VCdimension of the function class 'Hg 2 {ha,b; a, b [0, +oo)}, where ha), : R > {0, 1} is dened as 1, if |:1:| S a ha,b(m) = 1 if |33| 2 b 0 otherwise. 3. Consider the class 'Hl. Given a training data set 3, let h} be the outcome of the empirical risk minimization procedure restricted to class H1. How many training samples do we need (ignoring constants) to ensure that for any distribution of the data, D, the following event occurs with probability 1 5? L1)(h*S) 111161;? L901) 5 6. (Justify your answer in one sentence.) 4. Consider again the same class 7h. Recall that we dened hf'g to be the outcome of the empirical risk minimization procedure restricted to class 'Hl. How many training samples do we need (ignoring constants) to ensure that for any distribution of the data, D, the following event occurs with probability 1 6? L h* min L h