TensorFlow Reinforcement Learning Quick Start Guide Get up and Running with Training and Deploying Intelligent, Self-Learning Agents Using Python

• Main Author: Balakrishnan, Kaushik
• Format: eBook
• Language: English
• Published: Birmingham Packt Publishing, Limited 2019; Packt Publishing Limited; Packt Publishing
• Edition: 1
• Subjects: COM004000 COMPUTERS / Intelligence (AI) & Semantics; COMPUTERS / Data Modeling & Design; COMPUTERS / Machine Theory
• ISBN: 9781789533583; 1789533589
• DOI: 10.0000/9781789533446

This book is an essential guide for anyone interested in Reinforcement Learning. The book provides an actionable reference for Reinforcement Learning algorithms and their applications using TensorFlow and Python. It will help readers leverage the power of algorithms such as Deep Q-Network (DQN), Dee...

Table of Contents:

• Evaluating the performance of DDQN on Atari Breakout -- Understanding dueling network architectures -- Coding dueling network architecture and training it to play Atari Breakout -- Combining V and A to obtain Q -- Evaluating the performance of dueling architectures on Atari Breakout -- Understanding Rainbow networks -- DQN improvements -- Prioritized experience replay -- Multi-step learning -- Distributional RL -- Noisy nets -- Running a Rainbow network on Dopamine -- Rainbow using Dopamine -- Summary -- Questions -- Further reading -- Chapter 5: Deep Deterministic Policy Gradient -- Technical requirements -- Actor-Critic algorithms and policy gradients -- Policy gradient -- Deep Deterministic Policy Gradient -- Coding ddpg.py -- Coding AandC.py -- Coding TrainOrTest.py -- Coding replay_buffer.py -- Training and testing the DDPG on Pendulum-v0 -- Summary -- Questions -- Further reading -- Chapter 6: Asynchronous Methods - A3C and A2C -- Technical requirements -- The A3C algorithm -- Loss functions -- CartPole and LunarLander -- CartPole -- LunarLander -- The A3C algorithm applied to CartPole -- Coding cartpole.py -- Coding a3c.py -- The AC class -- The Worker() class -- Coding utils.py -- Training on CartPole -- The A3C algorithm applied to LunarLander -- Coding lunar.py -- Training on LunarLander -- The A2C algorithm -- Summary -- Questions -- Further reading -- Chapter 7: Trust Region Policy Optimization and Proximal Policy Optimization -- Technical requirements -- Learning TRPO -- TRPO equations -- Learning PPO -- PPO loss functions -- Using PPO to solve the MountainCar problem -- Coding the class_ppo.py file -- Coding train_test.py file -- Evaluating the performance -- Full throttle -- Random throttle -- Summary -- Questions -- Further reading -- Chapter 8: Deep RL Applied to Autonomous Driving -- Technical requirements -- Car driving simulators
• Learning to use TORCS -- State space -- Support files -- Training a DDPG agent to learn to drive -- Coding ddpg.py -- Coding AandC.py -- Coding TrainOrTest.py -- Training a PPO agent -- Summary -- Questions -- Further reading -- Assessment -- Chapter 1 -- Chapter 3 -- Chapter 4 -- Chapter 5 -- Chapter 6 -- Chapter 7 -- Chapter 8 -- Other Books You May Enjoy -- Index
• Cover -- Title Page -- Copyright and Credits -- Dedication -- About Packt -- Contributors -- Table of Contents -- Preface -- Chapter 1: Up and Running with Reinforcement Learning -- Why RL? -- Formulating the RL problem -- The relationship between an agent and its environment -- Defining the states of the agent -- Defining the actions of the agent -- Understanding policy, value, and advantage functions -- Identifying episodes -- Identifying reward functions and the concept of discounted rewards -- Rewards -- Learning the Markov decision process -- Defining the Bellman equation -- On-policy versus off-policy learning -- On-policy method -- Off-policy method -- Model-free and model-based training -- Algorithms covered in this book -- Summary -- Questions -- Further reading -- Chapter 2: Temporal Difference, SARSA, and Q-Learning -- Technical requirements -- Understanding TD learning -- Relation between the value functions and state -- Understanding SARSA and Q-Learning -- Learning SARSA -- Understanding Q-learning -- Cliff walking and grid world problems -- Cliff walking with SARSA -- Cliff walking with Q-learning -- Grid world with SARSA -- Summary -- Further reading -- Chapter 3: Deep Q-Network -- Technical requirements -- Learning the theory behind a DQN -- Understanding target networks -- Learning about replay buffer -- Getting introduced to the Atari environment -- Summary of Atari games -- Pong -- Breakout -- Space Invaders -- LunarLander -- The Arcade Learning Environment -- Coding a DQN in TensorFlow -- Using the model.py file -- Using the funcs.py file -- Using the dqn.py file -- Evaluating the performance of the DQN on Atari Breakout -- Summary -- Questions -- Further reading -- Chapter 4: Double DQN, Dueling Architectures, and Rainbow -- Technical requirements -- Understanding Double DQN -- Coding DDQN and training to play Atari Breakout
• TensorFlow Reinforcement Learning Quick Start Guide: Get up and running with training and deploying intelligent, self-learning agents using Python