All you need to know about Markov Decision processes, value- and policy-iteation as well as about Q learning approach
This course is about Reinforcement Learning. The first step is to talk about the mathematical background: we can use a Markov Decision Process as a model for reinforcement learning. We can solve the problem 3 ways: value-iteration, policy-iteration and Q-learning. Q-learning is a model free approach so it is state-of-the-art approach. It learns the optimal policy by interacting with the environment. So these are the topics:
Markov Decision Processes
value-iteration and policy-iteration
Q-learning fundamentals
pathfinding algorithms with Q-learning
Q-learning with neural networks
Understand reinforcement learning
Understand Markov Decision Processes
Understand value- and policy-iteration
7 Lectures
6 Lectures
4 Lectures
7 Lectures
5 Lectures
4 Lectures
1 Lectures
My name is Balazs Holczer. I am from Budapest, Hungary. I am qualified as a physicist. At the moment I am working as a simulation engineer at a multinational company. I have been interested in algorithms and data structures and its implementations especially in Java since university. Later on I got acquainted with machine learning techniques, artificial intelligence, numerical methods...
excellent course, well made snd explained!
I had liked some java example of deep reinforcement learning
I'm learning a lot.
This course was going in very great detail. Not easy. But thorough. Thank you for providing this series of 4 courses.
It's a really good course. I didn't realize that reinforcement learning is such powerful, especially when you combine it with deep learning.
excelente
This course make you more open mind
Good course to deep dive the reinforcement learning. Some improvements can be made
1. non-stationary cases
2. deep q-learning implementation (found one in machine learning course from the same instructor)
3. go beyond epsilon-greedy like UCB, thompson sampling
bom aprender e bosserver novidades
Your email address will not be published. Required fields are marked *
Sagaya K.
Very good