A single layer neural network with three neurons in the layer is to be designed for placement of students into programmes (Electrical, Computer, Management) at the Ashesi University. Three features of the applicants for consideration are: [age, aggregate, SHS background] where age feature: 17 24, aggregate feature: 6 36, and background feature: science = 1, business = 2, general arts = 3, home science = 4, and visual arts = 5. Assume the initial weight and bias of the network are: w0 = [w1, w2, w3] = [0.5, -0.2, 0.8], b = [b1, b2, b3] =[1.8, 0.6, -0.8), n = 0.2. (a) sketch the network architecture. (b) find the network weights and biases after one learning cycle when three training samples comprising a computer applicant with feature [19, 15, 2], electrical applicant with feature [18, 10, 1], and Management applicant with feature [19, 8, 5] are used for the training. Assume the network used a signum activation threshold function. Weight adjustment equation: Wk+1 = Wk + AW: where AW = ne(n)X for discrete learning function and AW= nenf (net)X for continuous learning function Bias adjustment equation : bk+1 = bk + Abk where Abk = ne(n)Xo, and Xo = 1 A single layer neural network with three neurons in the layer is to be designed for placement of students into programmes (Electrical, Computer, Management) at the Ashesi University. Three features of the applicants for consideration are: [age, aggregate, SHS background] where age feature: 17 24, aggregate feature: 6 36, and background feature: science = 1, business = 2, general arts = 3, home science = 4, and visual arts = 5. Assume the initial weight and bias of the network are: w0 = [w1, w2, w3] = [0.5, -0.2, 0.8], b = [b1, b2, b3] =[1.8, 0.6, -0.8), n = 0.2. (a) sketch the network architecture. (b) find the network weights and biases after one learning cycle when three training samples comprising a computer applicant with feature [19, 15, 2], electrical applicant with feature [18, 10, 1], and Management applicant with feature [19, 8, 5] are used for the training. Assume the network used a signum activation threshold function. Weight adjustment equation: Wk+1 = Wk + AW: where AW = ne(n)X for discrete learning function and AW= nenf (net)X for continuous learning function Bias adjustment equation : bk+1 = bk + Abk where Abk = ne(n)Xo, and Xo = 1