Share this: Google+

In this tutorial, you will discover step by step how an agent learns through training without teacher in unknown environment. Reinforcement learning is training paradigm for agents in which we have example of problems but we do not have the immediate exact answer. For playing a game, for instance, an agent will make series of decisions to move and only later will find out whether those decisions are right or wrong. Reinforcement learning paradigm is similar to real life of how we learn.

In this tutorial, you will find out part of reinforcement learning algorithm called
**
Q-learning
**
. Reinforcement learning algorithm has been widely used for many applications such as robotics, multi agent system, game, motion planning, navigation, and etc.

Instead of learning the theory of reinforcement that you can read it from many books and other web sites (see Resources for more references ), this tutorial will introduce the concept through simple but comprehensive numerical examples. If you purchase the e-book of this tutorial, you will also receive the companion worksheet and the matlab files .

Let us start the tutorial (clicks the topic below).

Modeling the Environment

Agent, State and Action Introduction

Q LearningQ Learning AlgorithmNumerical ExampleAnother Q learning Example: Tower of Hanoi

Q-Learning Solution for Tower of HanoiQ Learning using Matlab

Q Learning using MS Excel

Practice make perfect

Resources

Send your feedback and comments for this tutorial

**
Preferable reference for this tutorial is
**

Teknomo, Kardi. 2005. Q-Learning by Examples. http://people.revoledu.com/kardi/tutorial/ReinforcementLearning/index.html