Description
Deep Reinforcement Learning Hands-On by Maxim Lapan is a comprehensive guide to modern reinforcement learning (RL) methods. The book covers a range of topics, including deep Q-networks, value iteration, policy gradients, TRPO, and AlphaGo Zero. Readers will learn how to apply these techniques to solve complex real-world problems using deep learning (DL). The book provides an introduction to RL basics, enabling readers to code intelligent learning agents for various tasks. It includes practical examples such as implementing Q-learning in 'grid world' environments, training agents for stock trading, and defeating Atari arcade games using value iteration. Additionally, the book explores the latest deep RL research, including AI-driven chatbots. While some fluency in Python and basic DL approaches are assumed, no prior RL knowledge is required. The book was published by Packt Publishing in 2018, but a newer edition updated for 2020 is available, offering additional chapters on multi-agent methods and RL in robotics.
- CategoryEducation & Research
Book Details
Explore deep reinforcement learning from first principles
Evaluate high-profile RL methods like TRPO and PPO
Learn Markov decision processes as RL foundation
Implement Q-learning in 'grid world' environments
Train agents for stock trading using OpenAI Gym
Defeat Atari games using value iteration
Teach agents to play Connect4 with AlphaGo Zero
Discover AI-driven chatbot research
Who This Book Is For
- Stock trading agent
- Atari game strategies
- Connect4 gameplay
- AI-driven chatbots
- Grid world environments
