Some were even given strong agents as targets, caring about nothing else but beating an already successful strategy. One might attempt to achieve air superiority at all costs another to focus on teching up another to try various “cheese” attempts like worker rushes and the like. So the machine learning agent was essentially split into hundreds of versions of itself, each given a slightly different task or strategy. Hundreds of agents were spawned and pitted against each other.īecause StarCraft is such a complex game, it would be silly to think that there’s a single optimal strategy that works in all situations.
0 kommentar(er)
