1. The sample complexity of confidence. In lecture we saw that in order to learn a class of VC-dimension d with confidence 1- and error bounded by e, it is sufficient to find (d/e) examples were necessary to guarantee an error rate ofe with confidence 1-50 for some constant fo 0, In this problem we will extend this lower bound to include a dependence on when we wish to learn to arbitrarily high confidence 1-8. Now, using the lower bound we proved in lecture, conclude that for any algorithm to find a representation with error at most e with probability 1- 5 when the data is labeled by 1. The sample complexity of confidence. In lecture we saw that in order to learn a class of VC-dimension d with confidence 1- and error bounded by e, it is sufficient to find (d/e) examples were necessary to guarantee an error rate ofe with confidence 1-50 for some constant fo 0, In this problem we will extend this lower bound to include a dependence on when we wish to learn to arbitrarily high confidence 1-8. Now, using the lower bound we proved in lecture, conclude that for any algorithm to find a representation with error at most e with probability 1- 5 when the data is labeled by