Re:StarCoder and StarCoderBase: 15.5B parameter models with ...
by nerpderp82 on Tuesday May 16, 2023 @08:55AM
(#35961208)
This is why all the Alpha%wildcard% projects use self play, that along with a fitness function, or score maximizer generates their training data rather than have to show it records of games. You just let it play the game.
[ Parent
| on HN ]
Re:StarCoder and StarCoderBase: 15.5B parameter models with ...
by reaperman on Tuesday May 16, 2023 @11:06AM
(#35962867)
Not all of them. AlphaStar for StarCraft was largely supervised learning, to mimic top human players. Once it was sufficiently mimicking existing replays, then it switched to Reinforcement Learning.
That's why you see it do "nonsensical" things like destroying the "unbuildable plates" at the bottom of its natural ramp. This 100% wouldn't happen if it had only learned through self-play.