Descriptions
Modern Reinforcement Learning: Actor-Critical Agents. In this advanced course on deep reinforcement learning, you will learn how to implement Policy Gradient, Actor Critic, Deep Deterministic Policy Gradient (DDPG), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Soft Actor Critical (SAC) algorithms in a variety of challenging environments from the Open AI gym. There will be a strong emphasis on dealing with continuous action space environments, which will be of particular interest to those who want to pursue research in robot control using deep reinforcement learning. Rather than this being a spoon-fed-the-student course, here you will learn to read deep reinforcement learning research papers yourself and implement them from scratch. You will learn a repeatable framework for quickly implementing the algorithms in advanced research papers. Mastering the content of this course will represent a quantum leap in your skills as an artificial intelligence engineer and put you in a league of your own among students who rely on others to break down complex ideas for them.
Don’t worry if it’s been a while since your last intensive learning course, we’ll start with a brisk review of the core topics. And move straight on to programming our first agent: a blackjack playing artificial intelligence. From there, we’ll move on to teaching an agent to balance the cart pole using Q-learning. After learning the basics, things will move faster and we’ll jump right into introducing policy gradient methods. We’ll cover the REINFORCE algorithm and use it to teach an artificial intelligence to land on the moon in the lunar landing environment from the Open AI gym. Next, we’ll program the one-step Actor-Critical algorithm to defeat the lunar module again. Now that we’ve got the basics out of the way, we’ll move on to our more difficult projects: implementing research on deep reinforcement learning. We start with Deep Deterministic Policy Gradients (DDPG), an algorithm that teaches robots to excel at a variety of continuous control tasks. DDPG combines many of the advances of Deep Q Learning with traditional actor critique methods to achieve state-of-the-art results in environments with continuous action spaces.
What you will learn
- How to code policy gradient methods in PyTorch
- How to code Deep Deterministic Policy Gradients (DDPG) in PyTorch
- How to code Twin Delayed Deep Deterministic Policy Gradients (TD3) in PyTorch
- How to code actor critic algorithms in PyTorch
- How to implement cutting-edge artificial intelligence research in Python
Who is this course suitable for?
- Advanced students of Artificial Intelligence who want to carry out cutting-edge scientific research
Specificity of modern reinforcement learning: actor-critical agents
- Publisher : Udemy
- Teacher: Phil Tabor
- Language: English
- Level: Expert
- Number of courses: 58
- Duration: 8 hours and 10 minutes
Contents
Requirements
- Understanding analysis at university level
- Previous Reinforcement Learning courses
- Can program deep neural networks independently
Pictures
Sample clip
installation Guide
Extract the files and watch them with your favorite player
Subtitles: Not available
Quality: 720p
Download links
Password file(s): free download software
File size
2.91GB