Are you so sure? APM limits on the AI aren't necessary, which is a real insight FTA:
>In its games against TLO and MaNa, AlphaStar had an average APM of around 280, significantly lower than the professional players, although its actions may be more precise. This lower APM is, in part, because AlphaStar starts its training using replays and thus mimics the way humans play the game.
Where does it say the APM spiked during battles? If it's using it head-to-head with the human player, then that's bad, but how do you know it's not just herding workers and units individually?
I get the analogy but based on the distribution, 75% of the APM for alphastar is below the mean of APM for the human during gameplay. Following that analogy, both cars are on the racetrack, except one is applying acceleration with more precision - i.e. accelerating out of a turn rather than always accelerating. Capping acceleration wouldn't matter in this case.
The APM maxed out at 1500 at one point, far in excess of any human, especially considering many human actions are meaningless actions or spam clicks. The AI could win based on inhuman micro with stalkers and blink. I would be far more interested in seeing the agent with a fully realistic camera and fully realistic apm limit.
It does make a difference, because if the APM spike is during a head-on interaction with the human then it provides a competitive advantage, if it's randomly during another part of the match, then it doesn't provide a competitive advantage.
>In its games against TLO and MaNa, AlphaStar had an average APM of around 280, significantly lower than the professional players, although its actions may be more precise. This lower APM is, in part, because AlphaStar starts its training using replays and thus mimics the way humans play the game.
https://deepmind.com/blog/alphastar-mastering-real-time-stra...