0
$\begingroup$

my PPO training for a custom gymnasium environment resulted in following outcome. I would need some advice how to interpret the results and where to start activities to improve.

Thank you very much for all your support!

High-level info:

  • 60 iterations with 6000 timesteps
  • After 60 iterations, I had 149 episodes (1 episode always consist of 40 steps within episode)
  • learning rate was 0.001 (1e-3)

Learning stats

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

Hard to say much without knowing the specifics of the problem setting, reward function. Seems like trying out a couple of different hyper-parameters may help you get better performance, for example you can try reducing the learning rate, the loss looked volatile to me. Have you tried any other RL algorithms and compared the mean reward?

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.