Package: ReinforcementLearning 1.0.5

Introduction | The reinforcement learning problem | Policy learning | Experience replay | Setup of the ReinforcementLearning package | Installation | Package loading | Usage | Data preparation | Learning from pre-defined observations | Dynamic learning from an interactive environment function | Learning phase | General setup | Parameter configuration | Diagnostics | Working example 1: Gridworld | Problem definition | Defining an environment function | Learning an optimal policy | Evaluating policy learning | Applying a policy to unseen data | Updating an existing policy | Working example 2: Tic-Tac-Toe | Notes on performance | License