2. (20 points) Neural Networks. Consider a neural network, with Ni=2 input units, x= (x1,x2),Nh=3 hidden units and one output unit for binary classification: zjH=k=1NiWjkHxk+bjH,ujH={11ifzjH>0ifzjH0,j=1,,NhzO=k=1NhWkOukH+bO,y={10ifzO>0ifzO0. (a) (10 points) Suppose that WH=011111,bH=211 For each hidden unit j=1,2,3, draw the regions of inputs (x1,x2) where ujH=1. (b) (10 points) Find a vector of output weights WO and bias bO such that: - x=(0,0),(2,0),(1,2) are classified as y=0; and - x=(1,0.5) is classified as y=1