Reinforcement learning with Numpy ONLY - A gentle intro to the Bellman equation

preview_player
Показать описание
Can we give machines a growth mindset? 🤔 This video is a gentle introduction to reinforcement learning and the Bellman equation. We use Python to solve the system of Bellman equations numerically.

You can find the code that we develop in this video on my GitHub:

0:00 Intro
0:13 Basic elements of RL
1:05 Grid environement
1:50 Rewards
2:15 The goal of RL
3:29 What problem does RL solve?
3:51 Intro to the Bellman equation
4:54 Building the growth mindset!
5:50 Value function
8:44 Generalizing the value function -- Bellman equation
11:07 Linear system of Bellman equations
14:01 Solving the system of Bellman equations in Python!
15:17 Python code
17:37 Visualizing the solution
18:37 Outro

✨ If you'd like to support my efforts in creating open-source educational materials, you can buy me a croissant 🥐! Be sure to download your free Extras as a thank you for your support! 🚀
Рекомендации по теме