Book Image

Deep Reinforcement Learning with Python - Second Edition

By : Sudharsan Ravichandiran
Book Image

Deep Reinforcement Learning with Python - Second Edition

By: Sudharsan Ravichandiran

Overview of this book

With significant enhancements in the quality and quantity of algorithms in recent years, this second edition of Hands-On Reinforcement Learning with Python has been revamped into an example-rich guide to learning state-of-the-art reinforcement learning (RL) and deep RL algorithms with TensorFlow 2 and the OpenAI Gym toolkit. In addition to exploring RL basics and foundational concepts such as Bellman equation, Markov decision processes, and dynamic programming algorithms, this second edition dives deep into the full spectrum of value-based, policy-based, and actor-critic RL methods. It explores state-of-the-art algorithms such as DQN, TRPO, PPO and ACKTR, DDPG, TD3, and SAC in depth, demystifying the underlying math and demonstrating implementations through simple code examples. The book has several new chapters dedicated to new RL techniques, including distributional RL, imitation learning, inverse RL, and meta RL. You will learn to leverage stable baselines, an improvement of OpenAI’s baseline library, to effortlessly implement popular RL algorithms. The book concludes with an overview of promising approaches such as meta-learning and imagination augmented agents in research. By the end, you will become skilled in effectively employing RL and deep RL in your real-world projects.
Table of Contents (22 chapters)
18
Other Books You May Enjoy
19
Index

Summary

We started the chapter by understanding how to set up our machine by installing Anaconda and the Gym toolkit. We learned how to create a Gym environment using the gym.make() function. Later, we also explored how to obtain the state space of the environment using env.observation_space and the action space of the environment using env.action_space. We then learned how to obtain the transition probability and reward function of the environment using env.P. Following this, we also learned how to generate an episode using the Gym environment. We understood that in each step of the episode we select an action using the env.step() function.

We understood the classic control methods in the Gym environment. We learned about the continuous state space of the classic control environments and how they are stored in an array. We also learned how to balance a pole using a random agent. Later, we learned about interesting Atari game environments, and how Atari game environments are named in Gym, and then we explored their state space and action space. We also learned how to record the agent's gameplay using the wrapper class, and at the end of the chapter, we discovered other environments offered by Gym.

In the next chapter, we will learn how to find the optimal policy using two interesting algorithms called value iteration and policy iteration.