Contemplating Monopoly simulator—play until you get bored, then input game state and run Monte Carlo simulation to decide who was winning.
@peterseibel @mattknox ... for which, per your first tweet, Monte Carlo evaluation is the easiest thing. (For Go, anyway).
-
-
@avibryant@peterseibel@mattknox I still want to implement the bandit-based tree searches for games like this. -
@posco@avibryant@mattknox You mean bandits on top of Monte Carlo? I.e. weight the arms based on the results of each successive simulation? -
@peterseibel@posco@mattknox yes. Monte Carlo Tree Search. - 1 more reply
New conversation -
-
-
@avibryant@peterseibel@mattknox I would not say Go has few choices, though you are right about MCTSThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.