Chapter 11. Building Gaming Bot

In previous chapters, we covered applications that belong to the computer vision domain. In this chapter, we will be making a gaming bot. We will cover different approaches to build the gaming bot. These gaming bots can be used to play a variety of Atari games.

Let's do a quick recap of the past two years. Let's begin with 2015. A small London-based company called DeepMind published a research paper titled Playing Atari with Deep Reinforcement Learning, available at https://arxiv.org/abs/1312.5602 In this paper, they demonstrated how a computer can learn and play Atari 2600 video games. A computer can play the game just by observing the screen pixels. Our computer game agent (the computer game player) will receive rewards when the game score increases. The result presented in this paper is remarkable. The paper created a lot of buzz, and that was because each game has different scoring mechanisms and these games are designed in such a way that humans find it...