https://github.com/locuslab/ase
Analogous Safe-state Exploration (ASE) is an algorithm for provably safe and optimal exploration in MDPs with unknown, stochastic dynamics.
https://github.com/locuslab/ase
Last synced: about 1 year ago
JSON representation
Analogous Safe-state Exploration (ASE) is an algorithm for provably safe and optimal exploration in MDPs with unknown, stochastic dynamics.
- Host: GitHub
- URL: https://github.com/locuslab/ase
- Owner: locuslab
- Created: 2020-07-01T15:22:05.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2021-03-22T18:17:55.000Z (over 5 years ago)
- Last Synced: 2025-04-02T20:11:21.375Z (about 1 year ago)
- Language: Python
- Size: 104 KB
- Stars: 11
- Watchers: 4
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Analogous Safe-state Exploration
===
Building:
--------
To build, simply run `python setup.py build_ext --inplace`
Running:
--------
To run the experiments, run `main.py` with a YAML config file.
Example: `python main.py --config experiment_configs/unsafe_grid_world.yaml`