Reinforcement Learning in R

Abstract

This vignette gives an introduction to the ReinforcementLearning package, which allows one to perform model-free reinforcement in R . The implementation uses input data in the form of sample sequences consisting of states, actions and rewards. Based on such training examples, the package allows a reinforcement learning agent to learn an optimal policy that defines the best possible action in each state. In the following sections, we present multiple step-by-step examples to illustrate how to take advantage of the capabilities of the ReinforcementLearning package. Moreover, we present methods to customize the learning and action selection behavior of the agent. Main features of ReinforcementLearning include, but are not limited to: o Learning an optimal policy from a fixed set of a priori known transition samples