Loading...

How to Solve a Basic Reinforcement Learning Example | RL Hello World

3,383 views

Loading...

Loading...

Transcript

The interactive transcript could not be loaded.

Loading...

Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Streamed live on Jan 26, 2020

Useful timestamps:
0:24 Video Goal
0:50 Initial Q-Learning Explanation
7:30 Coding starts
35:10 Q&A
46:00 Agent can randomly solve maze
1:00:00 Agent scoring implemented
1:07:00 Setting up 3D visualizer
1:32:00 Detailed discussion of Q matrix, setting things up
1:52:00 Starting to set up QLearning class
2:16:00 Debugging issue in Q matrix update
2:46:00 Setting up larger mazes
3:01:11 Setting up explore vs exploit tradeoff
3:16:00 End of coding, Audience Questions and Discussion


In this recorded live stream I solve a basic reinforcement learning example, what you might consider a hello world for RL. Specifically I set up and solve an arbitrary maze using the Q Learning algorithm. The example is coded in Python and is done from scratch. I know the video is long but everything is covered at a basic level and should be very accessible.

SOME OF MY OTHER VIDOES:
○ Explaining RL to a baby: https://www.youtube.com/watch?v=GFIKt...
○ Coding on an Android: https://www.youtube.com/watch?v=bNx2d...
○ Learning programming language Julia: https://www.youtube.com/watch?v=TNoSh...
○ Making Python fast: https://www.youtube.com/watch?v=XW32l...

TWITTER: https://twitter.com/safijari

Loading...

Live chat replay is not available for this video.
Advertisement
When autoplay is enabled, a suggested video will automatically play next.

Up next


to add this to Watch Later

Add to

Loading playlists...