Book Image

Reinforcement Learning with TensorFlow

By : Sayon Dutta
Book Image

Reinforcement Learning with TensorFlow

By: Sayon Dutta

Overview of this book

Reinforcement learning (RL) allows you to develop smart, quick and self-learning systems in your business surroundings. It's an effective method for training learning agents and solving a variety of problems in Artificial Intelligence - from games, self-driving cars and robots, to enterprise applications such as data center energy saving (cooling data centers) and smart warehousing solutions. The book covers major advancements and successes achieved in deep reinforcement learning by synergizing deep neural network architectures with reinforcement learning. You'll also be introduced to the concept of reinforcement learning, its advantages and the reasons why it's gaining so much popularity. You'll explore MDPs, Monte Carlo tree searches, dynamic programming such as policy and value iteration, and temporal difference learning such as Q-learning and SARSA. You will use TensorFlow and OpenAI Gym to build simple neural network models that learn from their own actions. You will also see how reinforcement learning algorithms play a role in games, image processing and NLP. By the end of this book, you will have gained a firm understanding of what reinforcement learning is and understand how to put your knowledge to practical use by leveraging the power of TensorFlow and OpenAI Gym.
Table of Contents (21 chapters)
Title Page
Packt Upsell


In reinforcement learning, we want the Q-function Q(s,a) to predict the best action for a state s in order to maximize the future reward. The Q-function is estimated using Q-learning, which involves the process of updating the Q-function using Bellman equations through a series of iterations as follows:


Q(s,a) = Q value for the current state s and action a pair

 = learning rate of convergence

 = discounting factor of future rewards

Q(s',a') = Q value for the state action pair at the resultant state s' after action a was taken at state s

R = refers to immediate reward

 = future reward

In simpler cases, where state space and action space are discrete, Q-learning is implemented using a Q-table, where rows represent the states and columns represent the actions. 

Steps involved in Q-learning are as follows:

  1. Initialize Q-table randomly
  2. For each episode, perform the following steps:
    1. For the given state s, choose action a from the Q-table
    2. Perform action a
    3. Reward R and state s' is observed
    4. Update...