Book Image

Hands-On Intelligent Agents with OpenAI Gym

By : Palanisamy P
Book Image

Hands-On Intelligent Agents with OpenAI Gym

By: Palanisamy P

Overview of this book

Many real-world problems can be broken down into tasks that require a series of decisions to be made or actions to be taken. The ability to solve such tasks without a machine being programmed requires a machine to be artificially intelligent and capable of learning to adapt. This book is an easy-to-follow guide to implementing learning algorithms for machine software agents in order to solve discrete or continuous sequential decision making and control tasks. Hands-On Intelligent Agents with OpenAI Gym takes you through the process of building intelligent agent algorithms using deep reinforcement learning starting from the implementation of the building blocks for configuring, training, logging, visualizing, testing, and monitoring the agent. You will walk through the process of building intelligent agents from scratch to perform a variety of tasks. In the closing chapters, the book provides an overview of the latest learning environments and learning algorithms, along with pointers to more resources that will help you take your deep reinforcement learning skills to the next level.
Table of Contents (12 chapters)

Spaces in the Gym

We can see that each environment in the Gym is different. Every game environment under the Atari category is also different from the others. For example, in the case of the VideoPinball-v0 environment, the goal is to keep bouncing a ball with two paddles to collect points based on where the ball hits, and to make sure that the ball never falls below the paddles, whereas in the case of Alien-v0, which is another Atari game environment, the goal is to move through a maze (the rooms in a ship) collecting dots, which are equivalent to destroying the eggs of the alien. Aliens can be killed by collecting a pulsar dot and the reward/score increases when that happens. Do you see the variations in the games/environments? How do we know what types of actions are valid in a game?

In the VideoPinball environment, naturally, the actions are to move the paddles up or down...