When we use DDQN, we often use the target network in case our online network overestimates a value, but this doesn't make sense to me, because

  1. What happens if our target network is the one that overestimated a value, then we’d keep using that overestimated value

  2. Why can't we use both target network for selection and evaluation


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.