kumulant

UcbVSpec

@Serializable
@SerialName(value = "UcbV")
data class UcbVSpec(val zeta: Double = 1.2, val c: Double = 1.0, val priorMean: Double = 0.0, val priorWeight: Double = 0.02) : BanditPolicySpec<MomentsResult> (source)

Spec for UcbV.

Constructors

Link copied to clipboard
constructor(zeta: Double = 1.2, c: Double = 1.0, priorMean: Double = 0.0, priorWeight: Double = 0.02)

Properties

Link copied to clipboard
val c: Double

Bias-correction term scale.

Link copied to clipboard

Per-arm prior on the running reward mean.

Link copied to clipboard

Per-arm prior pseudo-count.

Link copied to clipboard

Variance-term scale.

Functions

Link copied to clipboard

Build a live BanditPolicy from its spec.

UcbVSpec

constructor(zeta: Double = 1.2, c: Double = 1.0, priorMean: Double = 0.0, priorWeight: Double = 0.02)(source)

c

Bias-correction term scale.

priorMean

Per-arm prior on the running reward mean.

priorWeight

Per-arm prior pseudo-count.

zeta

Variance-term scale.