qzed/irl-maxent
Maximum Entropy and Maximum Causal Entropy Inverse Reinforcement Learning Implementation in Python
This project helps machine learning researchers or roboticists understand the underlying reward function that explains observed expert behavior. It takes demonstrations of an optimal agent's actions in an environment and outputs a reward function that can then be used to train new agents. This is ideal for those working on reinforcement learning tasks where direct reward function design is difficult.
312 stars. No commits in the last 6 months. Available on PyPI.
Use this if you need to infer the reward function driving expert demonstrations in a simulated or real-world environment.
Not ideal if you already have a well-defined reward function or are not working with sequential decision-making problems.
Stars
312
Forks
63
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Apr 21, 2024
Commits (30d)
0
Dependencies
2
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/qzed/irl-maxent"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
DLR-RM/stable-baselines3
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
google-deepmind/dm_control
Google DeepMind's software stack for physics-based simulation and Reinforcement Learning...
Denys88/rl_games
RL implementations
pytorch/rl
A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.
yandexdataschool/Practical_RL
A course in reinforcement learning in the wild