Algorithm 6: mlEA-C PD -SFN

Input: G 0: Initial scale-free structure. GS: Population size. P c: Crossover rate. max: Levels in pyramids. gen_max: Iteration number. β: probability scale.

Output: G*: The optimized population structure.

Step 1: Initialize population through simple_edge_swapping(G0) and evaluate their cooperation level;

Step 2: EA population reproduce (Crossover operator). Select parents through roulette selection.

Step 3: Mutation (simple_edge_swapping) to the offspring and evaluate their cooperation level.

Step 4: Multilevel evolutionary operator performs upon both the parent and the offspring.

Step 5: If the iteration generation satisfies, output the best solution in EA population. Or else, select the best GS solutions in the EA population for the next generation and go to step 2.