ICLR 2024 Poster
Xiyao Wang · Ruijie Zheng · Yanchao Sun · Ruonan Jia · Wichayaporn Wongkamjan · Huazhe Xu · Furong Huang
We provide scripts to train and evaluate policies of different backbones (MBPO, DreamerV2, and DreamerV3) in separate folders.
For MBPO, in the MuJoCo environment, we implement COplanner based on mbpo_pytorch. In DeepMind Control Suite, we utilize MBRL-lib.
This repository is released under the MIT license. See LICENSE for additional details.
Our code is built upon mbpo_pytorch, MBRL-lib, DreamerV2-pytorch, and DreamerV3. We thank all these authors for their nicely open sourced code and their great contributions to the community.
If you find our work useful, please consider citing:
@article{wang2023coplanner,
title={COPlanner: Plan to Roll Out Conservatively but to Explore Optimistically for Model-Based RL},
author={Wang, Xiyao and Zheng, Ruijie and Sun, Yanchao and Jia, Ruonan and Wongkamjan, Wichayaporn and Xu, Huazhe and Huang, Furong},
journal={arXiv preprint arXiv:2310.07220},
year={2023}
}