Design a neural network t o solve the XOR problem i e the network should output 1 i f only one o f the two binary input variables i s 1 and 0 otherwise s e e i n Figure 1 a Use the hyperbolic tangent o r t a n h activation function i n all nodes F i g u r e 1 b which ranges i n 1 1 Figure 1 Note that A XOR B can b e expressed a s A O R B AND NOT A AND B a s illustrated below s e e i n Figure 2 Figure 3 I n the diagrams shown i n Figure 3 w e filled i n most o f the t a n h units parameters Fill i n the remaining parameters keeping i n mind that t a n h outputs 1 1 not 0 1 Note that w e need t o appropriately change the second layer t h e AND node t o take 1 1 a s inputs Also w e must add a n extra last layer t o convert the final output from 1 1 t o 0 1 Hint Assume that t a n h outputs 1 for any input x 2 1 for any input x 2 and 0 for x 0 1 Written Problem 1 2 5 points Design a neural network t o solve the XOR problem i e the network should output 1 i f only one o f the two binary input variables i s 1 and 0 otherwise s e e i n Figure 1 a Use the hyperbolic tangent o r t a n h activation function i n all nodes F i g u r e 1 b which ranges i n 1 1 Figure 1 Note that A XOR B can b e expressed a s A O R B AND NOT A AND B a s illustrated below s e e i n Figure 2 Figure 3 I n the diagrams shown i n Figure 3 w e filled i n most o f the t a n h units parameters Fill i n the remaining parameters keeping i n mind that t a n h outputs 1 1 not 0 1 Note that w e need t o appropriately change the second layer t h e AND node t o take 1 1 a s inputs Also w e must add a n extra last layer t o convert the final output from 1 1 t o 0 1 Hint Assume that t a n h outputs 1 for any input x 2 1 for any input x 2 and 0 for x 0