Problem 3 (30 points): Backpropagation Consider the following network, where r denotes output units, y denotes hidden units, and z denotes input units. t1 t2 t3 Output Ci Hidden yi Inputs Zk Consider: a) [15 points] The cross-entropy error for a single example is: E= ->(t; log(xi) + (1 - ti) log(1 - Ti)), where t is the target, and the logistic activation function is: 1 ri = 1te si where si = > y;wji, where wj denotes the weight of the edge between the jth hidden unit and the ith output unit