A visual guide on Reinforcement Learning – the 6 things that makes it “click”
In this video, I will give you the "big picture" that makes everything click when it comes to learning Reinforcement Learning.
The slides, animations, and side material are all available on my Patreon!
You can also read my blog article here:
https://towardsdatascience.com/the-handbook-of-reinforcement-learning-guide-to-the-foundational-questions/
Follow me on Twitter: https://x.com/neural_avb
To join our Patreon, visit: https://www.patreon.com/NeuralBreakdownwithAVB
We'll break down a simple framework of just 6 fundamental questions that EVERY RL algorithm must try to answer. By understanding these core problems, you'll be able to understand, compare, and analyze any RL system you encounter.
Along the way, we are learning about states and actions, environments and agents, value-based vs policy-based, Q-Learning, Policy gradients, Actor Critics, Advantages, Model-based RL, and so much more!
Members get access to everything behind-the-scenes that goes into producing my videos - including slides, docs, and code. Plus, it supports the channel in a big way and helps to pay my bills.
Timestamps:
0:00 - Intro
2:59 - Basics of RL
6:44 - What it can see, what it can do
9:03 - How it explores
11:43 - Models and Dynamics
13:50 - Evaluating states and Q-values
19:37 - TD Learning, MC Sampling
22:33 - Policy Gradients, Actor Critics
28:53 - Stability and Plasticity
32:00 - Outro
In this video, I will give you the “big picture” that makes everything click when it comes to learning Reinforcement Learning.
The slides, animations, and side material are all available on my Patreon!
You can also read my blog article here:
https://towardsdatascience.com/the-handbook-of-reinforcement-learning-guide-to-the-foundational-questions/
Follow me on Twitter: https://x.com/neural_avb
To join our Patreon, visit: https://www.patreon.com/NeuralBreakdownwithAVB
We’ll break down a simple framework of just 6 fundamental questions that EVERY RL algorithm must try to answer. By understanding these core problems, you’ll be able to understand, compare, and analyze any RL system you encounter.
Along the way, we are learning about states and actions, environments and agents, value-based vs policy-based, Q-Learning, Policy gradients, Actor Critics, Advantages, Model-based RL, and so much more!
Members get access to everything behind-the-scenes that goes into producing my videos – including slides, docs, and code. Plus, it supports the channel in a big way and helps to pay my bills.
Timestamps:
0:00 – Intro
2:59 – Basics of RL
6:44 – What it can see, what it can do
9:03 – How it explores
11:43 – Models and Dynamics
13:50 – Evaluating states and Q-values
19:37 – TD Learning, MC Sampling
22:33 – Policy Gradients, Actor Critics
28:53 – Stability and Plasticity
32:00 – Outro
