Multi-Armed Bandit Library v0.1

EpsilonDecreasingGambler Members

EpsilonDecreasingGambler overview

Public Instance Constructors

EpsilonDecreasingGambler ConstructorCreates a new epsilon decreasing gambler.

Public Instance Properties

EpsilonZeroGets the epsilon zero tuning parameter.
LeverCount (inherited from Gambler) Gets the number of levers (of arms) of the multi-armed bandit.
Score (inherited from Gambler)Sum of the observed values.

Public Instance Methods

Equals (inherited from Object)Determines whether the specified Object is equal to the current Object.
GetHashCode (inherited from Object)Serves as a hash function for a particular type, suitable for use in hashing algorithms and data structures like a hash table.
GetType (inherited from Object)Gets the Type of the current instance.
Observe (inherited from GamblerBase)Records of the reward brough by the specified lever.
PlayReturns the index of the pulled lever.
Reset (inherited from GamblerBase)Reset the counters of the GamblerBase.
ToString (inherited from Object)Returns a String that represents the current Object.

Protected Instance Fields

leverMeanSum (inherited from GamblerBase) The sum of the mean of the already observed levers.
leverSigmaSum (inherited from GamblerBase) The of the sigma of the levers already observed twice.
leverSquareMeanSum (inherited from GamblerBase) The sum of the square mean of the already observed levers.
observationCounts (inherited from GamblerBase) Contains the number of observations for each lever. The observation counts are indexed by the lever indices.
observedLeverCount (inherited from GamblerBase) Number of already observed levers.
rewardSquareSums (inherited from GamblerBase) Contains the sum of the squared of the collected rewards for each lever. The sums are indexed by the lever indices.
rewardSums (inherited from GamblerBase) Contains the sum of the collected rewards for each lever. The sums are indexed by the lever indices.
roundIndex (inherited from GamblerBase) Index of the round currently played when the method Play is called.
twiceObservedLeverCount (inherited from GamblerBase) Number of levers already observed twice.

Protected Instance Methods

Finalize (inherited from Object)Allows an Object to attempt to free resources and perform other cleanup operations before the Object is reclaimed by garbage collection.
LeverMean (inherited from GamblerBase)Gets the reward mean associated to the specified lever.
LeverSigma (inherited from GamblerBase) Gets the reward standard deviation associated to the specified lever.
MemberwiseClone (inherited from Object)Creates a shallow copy of the current Object.

See Also

EpsilonDecreasingGambler Class | Bandit.Stochastic Namespace