PoliciesMultiPlayers.CentralizedCycling module

CentralizedCycling: a multi-player policy which uses a centralized intelligence to affect users an offset, each one take an orthogonal arm based on (offset + t) % nbArms.

• It allows to have absolutely no collision, if there is more channels than users (always assumed).

• And it is perfectly fair on every run: each chosen arm is played successively by each player.

• Note that it is NOT affecting players on the best arms: it has no knowledge of the means of the arms, only of the number of arms nbArms.

class PoliciesMultiPlayers.CentralizedCycling.Cycling(nbArms, offset)

Bases: PoliciesMultiPlayers.BaseCentralizedPolicy.BaseCentralizedPolicy

Cycling: select an arm as (offset + t) % nbArms, with offset being decided by the CentralizedCycling multi-player policy.

__init__(nbArms, offset)

Cycling with an offset.

nbArms = None

Number of arms

offset = None

Offset

t = None

Internal time

__str__()

Return str(self).

startGame()

Nothing to do.

getReward(arm, reward)

Nothing to do.

choice()

Chose cycling arm.

__module__ = 'PoliciesMultiPlayers.CentralizedCycling'

class PoliciesMultiPlayers.CentralizedCycling.CentralizedCycling(nbPlayers, nbArms, lower=0.0, amplitude=1.0)

Bases: PoliciesMultiPlayers.BaseMPPolicy.BaseMPPolicy

CentralizedCycling: a multi-player policy which uses a centralize intelligence to affect users an offset, each one take an orthogonal arm based on (offset + t) % nbArms.

__init__(nbPlayers, nbArms, lower=0.0, amplitude=1.0)

• nbPlayers: number of players to create (in self._players).

• nbArms: number of arms.

Examples:

>>> import random; random.seed(0); import numpy as np; np.random.seed(0)
>>> s = CentralizedCycling(2, 3)
>>> [ child.choice() for child in s.children ]
[2, 1]
>>> [ child.choice() for child in s.children ]
[0, 2]
>>> [ child.choice() for child in s.children ]
[1, 0]
>>> [ child.choice() for child in s.children ]
[2, 1]

• To get a list of usable players, use s.children.

• Warning: s._players is for internal use

nbPlayers = None

Number of players

nbArms = None

Number of arms

children = None

List of children, fake algorithms

__str__()

Return str(self).

__module__ = 'PoliciesMultiPlayers.CentralizedCycling'

_printNbCollisions()

Print number of collisions.