Algorithm 4.
Mutate(Policy)
| Input: Policy to be mutated | |
| Data: L is the length of an individual’s string representation (i.e. length of decision variables vector); ℳ is the mutation rate. | |
| Output: Mutated policy | |
| 1 | begin |
| 2 | for i=1 to L do |
| 3 | x ← uniform random variate ∈ [0, 1]; |
| 4 | if x ≤ ℳ then |
| 5 | Policy[i] ← randomly selected PSA threshold ∈ {0, 0.5, …, 10, ∞}; |
| 6 | end |
| 7 | end |
| 8 | return Policy; |
| 9 | end |