Offline Goal Conditioned Reinforcement Learning with Temporal Distance Representations

---

Code for the paper "Offline Goal Conditioned Reinforcement Learning with Temporal Distance Representations".

This repository is a fork of OGBench. We add an implementation of the new Temporal Metric Distillation (TMD) algorithm, as well as a baseline implementation of Conservative Metric Distillation (CMD).

Citation

@article{myers2025offline,
  title={Offline Goal Conditioned Reinforcement Learning with Temporal Distance Representations},
  author={Myers, Vivek and Zheng, Bill Chunyuan and Eysenbach, Benjamin and Levine, Sergey},
  year={2025}
}