Skip to content
Code for generating options for planning and reinforcement learning
Branch: master
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.
options
simple_rl
.gitignore
README.md

README.md

Options for Planning and Reinforcement Learning.

Code for experiments on generating options for planning and reinforcement learning in our 2019 ICML papers:

Jinnai Y. Park JW, Abel D, Konidaris G. 2019. Discovering Options for Exploration by Minimizing Cover Time. Proc. 36th International Conference on Machine Learning.

Jinnai Y, Abel D, Hershkowitz E, Littman M, Konidaris G. 2019. Finding Options that Minimize Planning Time. Proc. 36th International Conference on Machine Learning

Dependencies

The code is written in Python 3. The code is dependent on numpy, scipy, and networkx. To solve MOMI optimally, ortools is required. simple_rl is a library for running RL experiments developed by David Abel. As I made a few tweaks to it, I'm putting the whole code in this repository here.

Directory

graph: approximation algorithms in graph algorithm literature. option_generation: option generation algorithms. experiments: Scripts to replicate experiments in papers.

Example

python3 options/experiments/planning_experiments.py
python3 options/experiments/rl_experiments.py

Author

Yuu Jinnai yuu_jinnai@brown.edu

Simple RL is developed by David Abel.

You can’t perform that action at this time.