Re:More on Dota 2
by pycal on Thursday August 17, 2017 @09:49AM
(#15036718)
Just a guess, but wouldn't this approach used in isolation result in a high potential for getting stuck at local optima? It seems to me that the success of this experiment is the result of interaction with players in different MMR brackets.
[ Parent
| on HN ]
Yes, it can. It's fairly common to keep a pool of learners which aren't identical for exactly this reason - either playing against prior snapshots, forking training, or lightly randomizing some components of the learner.