Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Adds some extra notes on RL training #1

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Conversation

LukeWood
Copy link

@LukeWood LukeWood commented Aug 9, 2020

I've recently been trying to take on RL again (bit of redemption after my
glaring defeat to box2d car racing 2 years ago!). In this pursuit I
came across some new analogies that were extremely useful to me in
creating a mental model for how some of these techniques work. Thought
they may be useful to some of your students.

This commit

  • adds an analogy for the purpose of the target network
  • emphasizes the reason experience replay works
  • adds a section on advanced RL techniques used to overcome sparse
    reward functions

I've recently been trying to take on RL again (bit of redemption after my
glaring defeat to box2d car racing 2 years ago!).  In this pursuit I
came across some new analogies that were extremely useful to me in
creating a mental model for how some of these techniques work.  Thought
they may be useful to some of your students.

This commit
- adds an analogy for the purpose of the target network
- emphasizes the _reason_ experience replay works
- adds a section on advanced RL techniques used to overcome sparse
reward functions
@LukeWood LukeWood requested a review from eclarson August 9, 2020 17:54
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant