Skip to content

Hwhitetooth/lirpg

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Learning Intrinsic Rewards for Policy Gradient

This repository is an implementation of On Learning Intrinsic Rewards for Policy Gradient Methods.

@article{zheng2018learning,
  title={On Learning Intrinsic Rewards for Policy Gradient Methods},
  author={Zheng, Zeyu and Oh, Junhyuk and Singh, Satinder},
  journal={arXiv preprint arXiv:1804.06459},
  year={2018}
}

Dependencies

This code is based on OpenAI baselines. In addtion, it requires the following:

  • Python 3.*
  • TensorFlow 1.7.0+

Training

To run A2C+LIRPG on Atari games:

python -m baselines.a2c.run_atari --env BreakoutNoFrameskip-v4

To run PPO+LIRPG on delayed Mujoco tasks:

python -m baselines.ppo2.run_mujoco --env Hopper-v2 --reward-freq 20

About

No description or website provided.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages