Deep Reinforcement Learning with Python

This book covers topics ranging from introduction to the latest advances in reinforcement learning with learning by coding. The theory is kept minimal with equipping the reader to assimilate and replicate the latest research in this field.This new edition focuses on the latest advances in Deep Reinforcement Learning. New agent environments ranging from games, and robotics to finance have been explained to help readers try different ways to apply reinforcement learning. It outlines the steps for using the code on multiple cloud systems and deploying models on online platforms such as Hugging Face Hub. A chapter on multi-agent reinforcement learning covers how multiple agents compete. This book contains a chapter on the widely used Deep RL algorithm, PPO (Proximal Policy Optimization). The reader will also understand how RLHF (Reinforcement Learning with Human feedback) has been used by chatbots, built using Large Language Models, e.g. ChatGPT to improve conversational capabilities.In the end, the reader will have a theoretical understanding and exposure to the most popular libraries in Deep Reinforcement Learning. The codes will be as Jupyter notebooks which could be run on Google Colab and similar other Deep Learning cloud platforms, allowing users to tailor the code to their own problems.What you will learn:Understand theoretical foundations of Reinforcement Learning and its most popular algorithms.Gain exposure of the most popular Python-based libraries in this field ranging from various environments for game play, robotics to even stock trading.Understand how to use highly optimized open-source libraries to train agents and ways to run your experiments on cloud Learn to use DRL (Deep Reinforcement Learning) specific frameworks and libraries in your own projects.Who This Book Is ForMachine learning developers and architects who want to stay ahead of the curve in the field of AI and deep learning.