In the experiments section of the HER paper, the authors mention that they added the square depth penetration term to the reward so as to penalize such behavior. Where do you get this quantity in the gym?
I've gone through the source code of the Fetch environments from OpenAI's Gym github page, but still can't find what I'm looking for. Is there some way to calculate it using some other variables?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.