Summary: G52PAS Planning and search 201011 mock exam
Answer 4 out of 6 questions.
1. (a) Explain A
search algorithm (give pseudocode) (10 marks)
(b) What is an admissible heuristic? Give an example.(5 marks)
(c) What is a consistent heuristic? Give an example. (5 marks)
(d) Give an example of a search problem and an admissible heuristic
where a graph search using A
is not optimal (does not return the
best solution). (5 marks)
2. (a) Explain the difference between statespace search algorithms and lo
cal search. What is the main reason for using local search? (5 marks)
(b) Explain hill climbing search (give pseudocode) (10 marks)
(c) Assume that the local search problem is to produce a natural number
which is greater than 41 and is divisible by 5. The only way to
generate successors is to add 1 or subtract 1 from the current state.
Design an objective function for this problem. (5 marks)
(d) Explain the problem of `getting trapped in local maxima' on the
example from part (b) (an example of a `local plateau' would do
as well). How does stochastic hill climbing attempts to solve this
