Fine-Tuning Parameters for Emergent Environments in Games Using Artificial Intelligence

Algorithm 1

Pseudocode for steepest ascent hill-climbing on fire related values.

int   tuneValueFire = 27;

real bestFitness;

for each cell{

   //tune for fire event

   real currentFitness = ⋯. //calculate the cell's fitness value

   While (tuneValueFire >0){

    Find the fitness when add the temperature value by tuneValueFire

    Find the fitness when subtract the temperature value by tuneValueFire

    Find the fitness when add the mass value by tuneValueFire

    Find the fitness when subtract the mass value by tuneValueFire

    Find the fitness when add the damage value by tuneValueFire

     …//try both add and subtract for all parameters of fire chromosomes

     …

    if (the best fitness value (compared to currentFitness) is found from the tuning

    trial results above){

       select a modification that causes such fitness

       commit changes according to the selected modification

     } else {

       tuneValueFire = tuneValueFire;

     }

   } // end while

} // end for each cell