Skip to content
Sparse Graphical Memory for Robust Planning
Jupyter Notebook Python
Branch: master
Clone or download

Latest commit

Latest commit f40826d Mar 20, 2020


Type Name Latest commit message Commit time
Failed to load latest commit information.
agents/thinned_fourrooms/FourRooms_coordinate_20steps-Dec-05-2019-12-58-28-PM arxiv code release Mar 13, 2020
notebooks arxiv code release Mar 13, 2020
sgm arxiv code release Mar 13, 2020
LICENSE Initial commit Mar 6, 2020 update bibtex format Mar 20, 2020
requirements.txt arxiv code release Mar 13, 2020

Sparse Graphical Memory (SGM)

Sparse Graphical Memory (SGM) is a data structure for reinforcement-learning agents to solve long-horizon, sparse-reward navigation tasks. This codebase is a TensorFlow implementation of SGM accompanying the paper Sparse Graphical Memory for Robust Planning.

To install:

  1. Create a new conda environment: conda create -n sgm python=3.6
  2. Activate the conda environment: conda activate sgm
  3. Install the requirements: pip install -r requirements.txt
  4. Install the package: pip install -e .

To launch the experiments:

  1. To launch SGM experiments, run all the cells in notebooks/SGM in Thinned FourRooms.ipynb
  2. To launch SoRB experiments, run all the cells in notebooks/SoRB in Thinned FourRooms.ipynb

To visualize the results:

  1. Call sgm/ with the directories of the logs produced by the experiments followed by the --returns_v_cleanup_steps flag, e.g., python sgm/ logs/thinned_fourrooms_sgm logs/thinned_fourrooms_sorb --returns_v_cleanup_steps
  2. Find the visualized results in the plots directory, e.g., in plots/returns_v_cleanup_steps

Commands you may find useful

Enable rendering of environments when there is no display, e.g., on a server: xvfb-run -s "-screen 0 1400x900x24" jupyter notebook

Create tunnel from local machine to remote machine: ssh -N -f -L localhost:8888:localhost:8888

Run Jupyter notebook with no time limit on cell execution and replace the notebook's contents with the new output: jupyter nbconvert --ExecutePreprocessor.timeout=-1 --execute --to notebook --inplace <notebook.ipynb>


This code was built upon the code released by Eysenbach et al. under the Apache License, Version 2.0.


  Author = {Laskin, Michael and Emmons,
  Scott and Jain, Ajay and Kurutach, Thanard
  and Abbeel, Pieter and Pathak, Deepak},
  Title = {Sparse Graphical
  Memory for Robust Planning},
  Booktitle = {arXiv preprint arXiv:2003.06417},
  Year = {2020}
You can’t perform that action at this time.