(a) The diagram shown represents an artificial neural network. [Figure 9.1]

Mark scheme

Marking points from the official mark scheme (Oct/Nov 2021). The mark or mark type is shown in brackets.

1. [1] One mark for correct statement (Max 1)
2. • Enables deep learning to take place
3. • Where the problem you are trying to solve has a higher level of complexity it requires more layers to solve
4. • To enable the neural network to learn and make decisions on its own
5. • To improve the accuracy of the result.
6. [4] One mark for each correct marking point (Max 4)
7. • Artificial neural networks are intended to replicate the way human brains work
8. • Weights / values are assigned for each connection between nodes
9. • The data are input at the input layer and are passed into the system
10. • They are analysed at each subsequent (hidden) layer where characteristics are extracted / outputs are calculated
11. • ... this process of training / learning is repeated many times to achieve optimum outputs // reinforcement learning takes place
12. • Decisions can be made without being specifically programmed
13. • The deep learning net will have created complex feature detectors
14. • The output layer provides the results
15. • Back propagation (of errors) will be used to correct any errors that have been made.
16. [5] One mark for each correct calculation as follows (Max 4)
17. • Node B (from Home) (Line 3 in table)
18. • Node C (from Home) (Line 4 in table)
19. • Node B and Node E (from A) (Lines 5 and 6 in table)
20. • Node F and Node School (from E) (Lines 7 and 8 in table)
21. • Node School (from F) (Line 9 in table)
22. One mark for correct path (Max 1):
23. • Home ⇨ A ⇨ E ⇨ F ⇨ School
24. Node Cost from Home Node (g) Heuristic (h) Total (f = g + h)
25. Home 0 14 14
26. A 1 10 11
27. B 5 7 12
28. C 4 9 13
29. B 1+3=4 7 11
30. E 1+6=7 3 10
31. F 7+1=8 3 11
32. School 7 + 5 = 12 0 12
33. School 8+3=11 0 11