Policy Optimization with Demonstrations (POfD)

This repository is a reimplementation of Policy Optimization with Demonstrations (ICML 2018).

Dependencies

This code is highly based on OpenAI baselines gail.

To run POfD on delayed Mujoco tasks:

python baselines/gail/run_mujoco.py --env_id Hopper-v1 --reward-freq 10 --num_epochs 1000

Name		Name	Last commit message	Last commit date
Latest commit History 10 Commits
baselines		baselines
data		data
LICENSE		LICENSE
README.md		README.md
run_ant.sh		run_ant.sh
run_halfcheetah.sh		run_halfcheetah.sh
run_hopper.sh		run_hopper.sh
run_humanoid.sh		run_humanoid.sh
run_walker2d.sh		run_walker2d.sh
setup.py		setup.py