please solve (iii) same example 2

Example 2: Consider the simple network below: Assume that the neurons have a Sigmoid activation function and Perform a forward pass on the network Perform a reserse pass (training) once (target =0.5 ). Perform a further forward pass and commeat on the result. Answer: (i) Input to top neuren =(0.350.1)+(0.90.8)=0.755. Out =0.68. Input to bottom neuron =(0.90.6)+(0.350.4)=0.68.0ut=0.6637. Input to final neuron =(0.30.68)+(0.90.6637)=0.80133,0 ut = 0.69. (iii) Output crror ^=(t0)(10)e=(0.50.69)(10.69)0.69=0.0406. New weights for output layer w1+=w11+(6x innut )=0.3+(0.04060.68)=0.272392. w2+=w2+( input )=0.9+(0.04060.66.37)=0.87305. Errors for hidden layers: 61=6w1=0.04060.272392(10)0=2.406103 82=w2=0.04060.87305(10)0=7.916103 New hidden layer weights: w3+=0.1+(2.4061030.35)=0.09916.w4+=0.8+(2.4061030.9)=0.7978w5+=0.4+(7.9161030.35)=0.3972.w6+=0.6+(7.9161030.9)=0.5928 (iii) Old error was 0.19. New error is 0.18205. Therefore error has reduced