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

implement REINFORCE++ algorithm #228

Merged
merged 4 commits into from
Feb 9, 2025

Conversation

4332001876
Copy link
Contributor

We have implemented the REINFORCE++ algorithm.

To use it, specify the parameter algorithm.adv_estimator=reinforce_plus_plus.

Preliminary performance evaluations were conducted within the Unakar/Logic-RL project, a reproduction of DeepSeek R1 Zero on the 2K Tiny Logic Puzzle Dataset. Results indicate that our REINFORCE++ implementation exhibits performance and training stability comparable to, or potentially exceeding, that of PPO and GRPO.

Related issue: #68

@vermouth1992
Copy link
Collaborator

Could you perform formatting according to the readme?

@4332001876
Copy link
Contributor Author

Modifications made according to your feedback.

@vermouth1992
Copy link
Collaborator

The formatting CI still fails.

@4332001876
Copy link
Contributor Author

Resolved formatting issues by running the script; sorry for the initial mistake.

@vermouth1992 vermouth1992 merged commit bdb50ac into volcengine:main Feb 9, 2025
11 checks passed
@PeterSH6 PeterSH6 mentioned this pull request Feb 9, 2025
4 tasks
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.

3 participants