VC-Example2Question 1Consider the case where the examples to a learning task are given as pairs of (not necessarily Boolean)numbers (x1, x2), labeled as positive or negative.Part AIt is known that all positive examples and none of the negative examples satisfy:x1= a, AND 0 ≤ x2≤ bHere are a few training examples for the case in which a = 3, b = 1:x1x2label3 1 positive3 0.5 positive3 0 positive0 0.5 negative3 2 negative5 6 negativeA1Select the most appropriate learning algorithm for this task among the following choices:1. ID3.2. A perceptron implemented with a sigmoid unit, with the following functions as input: φ1= 1, φ2= x1,φ3= x2.3. A neural network with the following functions as input: φ1= 1, φ2= x1, φ3= x2, with one hidden layerand with as many hidden layer nodes as needed.4. A neural network with the following functions as input: φ1= 1, φ2= x1, φ3= x2, with two hidden layersand with as many hidden layer nodes as needed.5. Naive Bayesian.6. Nearest Neighbor.Answer:1 / 2 / 3/ 4 / 5 / 6A2Your answers to this part should not depend on your answer to A1. Assume that a learning algorithmcapable of producing a hypothesis consistent with all training examples is available. In each of the followingcases compute how many randomly chosen training examples are needed to guarantee with confidence of atleast 90% that at least 95% of randomly selected test examples are answered correctly. Specify the formulayou use for the computation, and what is the value of each of the variables in the formula.1. The value of a is one of the following: 1, 1.5, 2, 2.5, 3, 3.5.The value of b is one of the following: 1, 1.5, 2, 2.5, 3, 3.5.Answer: The number of training examples should be at least .The formula used:••The variables in the formula have the values:••2. a, b are real numbers.Answer: The number of training examples should be at least .The formula used:••The variables in the formula have the
View Full Document