1-State whether true or false with a brief justification (No points will be given for random guessing without justification)

a.Given a dataset with noisy examples (i.e., it there are cases where for 2 examples, the feature values match but the class value does not), the training error for the ID3 algorithm is always guaranteed to be 0. b.Given a dataset with no noisy examples, Nave Bayes is guaranteed to have a training error equal to 0. c.Given any dataset, the leave-one-out cross validation error of 1-nearest neighbor is always equal to 0. d.( ) ( ) can be represented by a perceptron e.During the testing phase (making predictions given a trained classifier), a K-Nearest Neighbor classifier is typically far more efficient in terms of computation as compared to a Neural Network based classifier. f.A neural network with arbitrary number of hidden layers can represent every function that a decision tree can represent.