PoliciesMultiPlayers.CentralizedIMP module

CentralizedIMP: a multi-player policy where ONE policy is used by a centralized agent; asking the policy to select nbPlayers arms at each step, using an hybrid strategy: choose nb-1 arms with maximal empirical averages, then 1 arm with maximal index. Cf. algorithm IMP-TS [Komiyama, Honda, Nakagawa, 2016, arXiv 1506.00779].

class PoliciesMultiPlayers.CentralizedIMP.CentralizedIMP(nbPlayers, nbArms, playerAlgo, uniformAllocation=False, *args, **kwargs)

Bases: PoliciesMultiPlayers.CentralizedMultiplePlay.CentralizedMultiplePlay

CentralizedIMP: a multi-player policy where ONE policy is used by a centralized agent; asking the policy to select nbPlayers arms at each step, using an hybrid strategy: choose nb-1 arms with maximal empirical averages, then 1 arm with maximal index. Cf. algorithm IMP-TS [Komiyama, Honda, Nakagawa, 2016, arXiv 1506.00779].

_choice_one(playerId)

Use choiceIMP for each player.

__module__ = 'PoliciesMultiPlayers.CentralizedIMP'