Code accompanying the paper "Evolution-Guided Policy Gradients in Reinforcement Learning" accepted at NIPS 2018
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
core
README.md
run_erl.py

README.md

erl_paper_nips18

Code accompanying the paper "Evolution-Guided Policy Gradients in Reinforcement Learning" accepted at NIPS 2018

Dependencies

Python 3.5.6
Pytorch 0.3.1.post3
Numpy 1.15.2
Fastrand from https://github.com/lemire/fastrand
Gym 0.10.5
Mujoco-py v1.50.1.59

To Run

python run_erl.py -env $ENV_NAME$

ENVS TESTED

'Hopper-v2'
'HalfCheetah-v2'
'Swimmer-v2'
'Ant-v2'
'Walker2d-v2'
'Reacher-v2'