Background process

• Population size = 1000 agents
  

  training set = 25 agents:

  {

  15/1 fittest agents from foreground 10/24 fittest agents from background

  
  
• Iteration:
  1. All 1000 play against each 25 in training set
  2. Fitness evaluated
  3. Best half mate and generate new bottom half
  4. Best 10/24 (with fitness sharing) go to training set
• Every foreground iteration:
  1. Population is reset / seeded with best 100
  2. New 15/1 best are/is received