You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There seems to be some serious issue inside the monte carlo tree.
Actually at the current point, the playing strength of a MCTS with forced maximum treedepth of 1 is stronger than any other configuration, which makes no sense.
It may be related to switching player and reversing winrate stuff?
Needs to be investigated.
Maybe it would make sense to get some kind of best move->best response->... - list for debugging purposes
The text was updated successfully, but these errors were encountered:
There seems to be some serious issue inside the monte carlo tree.
Actually at the current point, the playing strength of a MCTS with forced maximum treedepth of 1 is stronger than any other configuration, which makes no sense.
It may be related to switching player and reversing winrate stuff?
Needs to be investigated.
Maybe it would make sense to get some kind of best move->best response->... - list for debugging purposes
The text was updated successfully, but these errors were encountered: