luanshaotong / LESSON

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Learning Subgoal Representations with Slow Dynamics

We propose a slowness objective to effectively learn the subgoal representation for goal-conditioned hierarchical reinforcement learning. Our paper is accepted by ICLR 2021.

The python dependencies are as follows.

Run the codes with python train_hier_sac.py. The tensorboard files are saved in the runs folder and the trained models are saved in the saved_models folder.

ezoic increase your site revenue

About


Languages

Language:Python 100.0%