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

RL4J: Weighted Double Q Learning #5275

Open
AlexDBlack opened this issue May 21, 2018 · 0 comments

Comments

Projects
None yet
1 participant
@AlexDBlack
Copy link
Contributor

commented May 21, 2018

From @tom-adsfund on May 12, 2018 3:5

Double Q Learning (greatly) improved on Q Learning to prevent inevitable upward bias.

But DQL has its own (downward) bias issues, which is where WDQL comes in.

Paper: https://www.ijcai.org/proceedings/2017/0483.pdf

Has been used recently in multi agent cooperative reinforcement with success: https://arxiv.org/pdf/1802.08534.pdf

Copied from original issue: deeplearning4j/rl4j#95

@AlexDBlack AlexDBlack changed the title Weighted Double Q Learning RL4J: Weighted Double Q Learning May 21, 2018

@AlexDBlack AlexDBlack added the RL4J label May 21, 2018

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.