imitation provides open-source implementations of imitation and reward learning algorithms in PyTorch. We include three inverse reinforcement learning (IRL) algorithms, three imitation learning algorithms and a preference comparison algorithm. The implementations have been benchmarked against previous results, and automated tests cover 98% of the code. Moreover, the algorithms are implemented in a modular fashion, making it simple to develop novel algorithms in the framework. Our source code, including documentation and examples, is available at https://github.com/HumanCompatibleAI/imitation
@article{arxiv.2211.11972,
title = {imitation: Clean Imitation Learning Implementations},
author = {Adam Gleave and Mohammad Taufeeque and Juan Rocamonde and Erik Jenner and Steven H. Wang and Sam Toyer and Maximilian Ernestus and Nora Belrose and Scott Emmons and Stuart Russell},
journal= {arXiv preprint arXiv:2211.11972},
year = {2022}
}