EpsilonGreedyPolicy: Epsilon Greedy Policy

Description Arguments Usage Examples

Description

Epsilon Greedy Policy

Arguments

epsilon

[numeric(1) in [0, 1]]
Ratio of random exploration in epsilon-greedy action selection.

Usage

makePolicy("epsilon.greedy", epsilon = 0.1)
makePolicy("greedy")

Examples

1
policy = makePolicy("epsilon.greedy", epsilon = 0.1)

markusdumke/reinforcelearn documentation built on May 31, 2019, 8:48 p.m.