Skip to content

[ECAI-2025] SPOWL: A JAX-based Safe RL framework that adaptively combines planning and policy learning with dynamic safety thresholds.

License

Notifications You must be signed in to change notification settings

AIRI-Institute/SPOWL

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Safe Planning and Policy Optimization via World Model Learning

Requirements

Installation

Get started with SPOWL:

  1. Create a conda environment:

    conda create -n spowl python==3.10
  2. Activate the environment:

    conda activate spowl
  3. Install Safety Gymnasium

    wget https://github.com/PKU-Alignment/safety-gymnasium/archive/refs/heads/main.zip
    unzip main.zip
    rm -rf main.zip
    pip install -e safety-gymnasium-main
  4. Install jax:

    pip install --no-cache-dir --upgrade pip
    pip install --no-cache-dir --upgrade "jax[cuda12]"
  5. Install other requirements:

    pip install --no-cache-dir hydra-core tabulate wandb tqdm moviepy equinox optax
  6. Install for 'osmesa':

    conda install -c conda-forge mesalib
  7. Fix dependencies:

    pip install --no-cache-dir gymnasium-robotics==1.2.3 numpy==1.25.0

Usage

Coming soon

SPOWL in some tasks

Point Goal 1

Point Goal 1

Point Goal 2

Point Goal 2

Point Button 1

Point Button 1

Point Push 1

Point Push 1

Car Goal 1

Car Goal 1

Doggo Goal 1

Doggo Goal 1

Ant Goal 1

Ant Goal 1

Citation

If you use SPOWL in your research, please cite:

@article{latyshev2025spowl,
  title={Safe Planning and Policy Optimization via World Model Learning},
  author={Latyshev, Artem and Gorbov, Gregory and Panov, Aleksandr I.},
  journal={arXiv preprint arXiv:2506.04828},
  year={2025}
}

About

[ECAI-2025] SPOWL: A JAX-based Safe RL framework that adaptively combines planning and policy learning with dynamic safety thresholds.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published