google-deepmind / google-deepmind/reverb
Update priority
Open
- Dominant language
- C++
- Stars
- 793
- Forks
- 113
- PR merge metrics
- No merged PRs in 30d
Description
Since I followed PER which is described in [this paper](https://arxiv.org/pdf/1511.05952.pdf), I believe that there must exist a way to update priority every time we compute TD error on every minibatch. However, I carefully looked at all your examples but it seems no function like this existing.
Contributor guide
Assessment
This issue has not been assessed yet.