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.

Sign up for GitHub

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

Jump to bottom

PPO ratio is incorrectly calculated #2

Open
deketh opened this issue Mar 27, 2024 · 0 comments
Open

PPO ratio is incorrectly calculated #2

deketh opened this issue Mar 27, 2024 · 0 comments
Assignees
@deketh

Comments

@deketh
Copy link
Owner

deketh commented Mar 27, 2024

Currnetly, continuous PPO ratio is calculated as ratios = new_actor_log_probs/(actor_log_probs+EPSILON) this is derived from previous PPO algorithm ratio = actor_probs/(old_actor_probs+EPSILON) and that made sense as they were not log probabilites.

The line should actually be something like ratios = exp(new_actor_log_probs - actor_log_probs)
@deketh deketh self-assigned this Mar 27, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@deketh deketh

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@deketh