Machine Learning Course: Reinforcement of Learning — Full Tutorial
This is NOT the AI technology used to beat GO, DOTA, or Chess. This means that you are not only generating words based on billions of sentences but also planning actions to defeat real game opponents. It’s also free.
Machine learning includes the area of reinforcement learning, which involves taking correct action in order to maximize rewards. This full tutorial will give you a solid grounding in the core reinforcement learning topics.
The course covers SARSA, Q learning and SARSA double learning. It also includes deep Q learning. These algorithms can be used in many different environments, such as space invaders and breakout. Tensorflow, PyTorch and PyTorch are used for the deep learning part.
The course starts with modern algorithms such as policy gradients and deep q-learning, and then demonstrates the power and effectiveness of reinforcement learning.
The course then teaches the basic concepts that underlie all algorithms for reinforcement learning. The course illustrates these concepts by coding some algorithms that were developed before deep learning but remain fundamental to cutting-edge technology. They are then studied in more traditional OpenAI environments, such as the cart pole problems.
Code: https://github.com/philtabor/Youtube-Code-Repository/tree/ma…ntLearning.
Course Contents
(00:00:00) Intro.
(00:01:30) Intro to Deep Q Learning.
How to code Deep Q learning in Tensorflow.
Deep Q Learning With Pytorch: Part 1: The Q Network.
Deep Q Learning Part 2: Coding Agent.
Deep Q Learning part with Pytorch (01:28.54).
01:46:39 Intro to Gradients of Policy 3: Coding the Main Loop
How to beat Lunar Lander using policy gradients (01:55)
How to beat Space Invaders using policy gradients.
How to create your own Reinforcement learning environment Part 1.
How to create your own Reinforcement learning environment Part 2.
Fundamentals of Reinforcement learning (03:08-20)
(03:17:09) Markov Decision Processes.
The Explore Exploit dilemma (03:23.02).
SARSA
Double Q Learning: Reinforcement learning in the Open AI Gym.
(03:54:07) Conclusion.
Machine Learning Course with Phil. Check out his YouTube channel: https://www.youtube.com/channel/UC58v9cLitc8VaCjrcKyAbrw.
Source:
Leave a Reply