Deep training with reinforcements on Python. Openai Gym and Tensorflow for pro

Deep training with reinforcement (Reinforced learning) is the most popular and promising direction of artificial intelligence.
Practical study of RL on Python will help to master not only basic, but also advanced algorithms of deep learning with reinforcement.
You will start with the basic principles of training with reinforcement, Openai Gym and Tensorflow, get acquainted with Markov chains, Monte Carlo and Dynamic programming, so that "terrible" abbreviations DQN, DRQN, A3C, PPO and TRPO will soon stop scaring you . You will learn about agents who learn on human preferences, DQFD, Her and many other recent RL achievements.
After reading the book, you will gain knowledge and experience necessary for the implementation of training with reinforcement and deep training with reinforcement in real projects, and enter the world of artificial intelligence.
In this book you:
- get acquainted with the basics of methods, algorithms and elements RL
- Teach an agent using Opena Gym and Tensorflow
- Master the Markov decision-making processes, Bellman"s optimality and TD training
- Learn to solve the problems of multi-armed bandits
- You will master the algorithms of deep learning, such as RNN, LSTM and CNN
- Create intellectual agents using the DRQN ​​algorithm that can play Doom
- Using DDPG, teach agents to play Lunar Lander
- Send the agent to the car races using the DQN method