GitHub·Code·MIT

initial-h/CEER

Replay Memory as An Empirical MDP: Combining Conservative Estimation with Experience Replay. ICLR 2023

4 stars·Updated Apr 2026
Open on GitHub

Authors

Repository owner

Linked papers

Code

Replay Memory as An Empirical MDP: Combining Conservative Estimation with Experience Replay

iclr · 2023

Why Lacuna links this resource
Repository names this paper

## Replay Memory as An Empirical MDP: Combining Conservative Estimation with Experience Replay ![overview](https://github.com/initial-h/ceer/blob/main/pic/overview.png) ## Overview - PyTorch implementation of Conservative Estimation with Experience Replay ([CEER](https://openreview.net/forum?id=SjzFVSJUt8S)). - Method is tested on [Sokoban](https://github.com/mpSchrader/gym-sokoban), [Minigrid](https://github.com/Farama-Foundation/Minigrid) and [MinAtar](https://github.com/kenjyoung/MinAtar) env

README:focal_exact_title·Open evidence