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

Soft Actor-Critic

The algorithm for Soft Actor-Critic (SAC) is given as follows:

  1. Initialize the main value network parameter , the Q network parameters and , and the actor network parameter
  2. Initialize the target value network by just copying the main value network parameter
  3. Initialize the replay buffer
  4. For N number of episodes, repeat step 5
  5. For each step in the episode, that is, for t = 0, . . ., T – 1
    1. Select action a based on the policy , that is,
    2. Perform the selected action a, move to the next state , get the reward r, and store the transition information in the replay buffer
    3. Randomly sample a minibatch of K transitions from the replay buffer
    4. Compute target state value
    5. Compute the loss of value network and update the parameter using gradient descent,
    6. Compute the target Q value
    7. Compute the loss of the Q networks and update the parameter using gradient descent...