Skip to content

Code for "Learning Exploration Strategies for Model Agnostic Meta-Reinforcement Learning", ICML AMTL workshop

License

Notifications You must be signed in to change notification settings

swami1995/exp_maml

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Learning Exploration Strategies for Model Agnostic Meta-Reinforcement Learning

Switch to branch another_sparse_branch_ppo to run the experiments on our model. The code in master is used to run experiments on the vanilla maml-trpo model.

Getting started

To avoid any conflict with your existing Python setup, and to keep this project self-contained, it is suggested to work in a virtual environment with virtualenv. To install virtualenv:

pip install --upgrade virtualenv

Create a virtual environment, activate it and install the requirements in requirements.txt.

virtualenv venv
source venv/bin/activate
pip install -r requirements.txt

Usage

You can use the main.py script in order to run reinforcement learning experiments. This script was tested with Python 3.5.

python main.py --env-name HalfCheetahRandVelEnv-v1 --fast-batch-size 20 --meta-batch-size 40 --output-folder hcv-1 --num-workers 16 --embed-size 32  --exp-lr 7e-4 --baseline-type nn --nonlinearity tanh --num-layers-pre 1 --hidden-size 64 --seed 0

References

A huge part of this implementation is borrowed from the MAML implementation of tristandeleu/pytorch-maml-rl in Pytorch. A huge thanks to them for open-sourcing their implementation.

About

Code for "Learning Exploration Strategies for Model Agnostic Meta-Reinforcement Learning", ICML AMTL workshop

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages