EpsilonGreedyPolicy

GreedyPolicy

<p>[<code>numeric(1) in [0, 1]</code>] 
Ratio of random exploration in epsilon-greedy action selection.</p>

epsilon

Implements reinforcement learning environments and algorithms as described in Sutton & Barto (1998, ISBN:0262193981).
The Q-Learning algorithm can be used with function approximation,
eligibility traces (Singh & Sutton (1996) <doi:10.1007/BF00114726>)
and experience replay (Mnih et al. (2013) <arXiv:1312.5602>).

Markus Dumke

reinforcelearn

Reinforcement Learning

EpsilonGreedyPolicy function

<p><code>makePolicy("epsilon.greedy", epsilon = 0.1)</code> 
<code>makePolicy("greedy")</code></p>

EpsilonGreedyPolicy: Epsilon Greedy Policy

Description

Arguments

Usage

Examples