Official codebase for Contrastive Energy Prediction for Exact Energy-Guided Diffusion Sampling in Offline Reinforcement Learning. Contains scripts to reproduce experiments.

Cheng Lu*, Huayu Chen*, Jianfei Chen†, Hang Su, Chongxuan Li, Jun Zhu†

*equal contribution, †equal advising

We provide code in two sub-directories: Offline_RL_2D containing code for example toy 2D and offline RL experiments and images containing code (based on implementations of openai/guided-diffusion) for Imagenet experiments. See corresponding READMEs in each folder for instructions; scripts should be run from the respective directories.

Please cite our paper as:

@article{lu2023contrastive,
  title={Contrastive Energy Prediction for Exact Energy-Guided Diffusion Sampling in Offline Reinforcement Learning},
  author={Lu, Cheng and Chen, Huayu and Chen, Jianfei and Su, Hang and Li, Chongxuan and Zhu, Jun},
  journal={arXiv preprint arXiv:2304.12824},
  year={2023}
}

MIT