Automata Learning from Preference and Equivalence Queries

The project website can be found at https://eric-hsiung.github.io/remap.

This is the experimental code repository for the paper Automata Learning from Preference and Equivalence Queries (CAV 2025).

It contains an implementation of the REMAP algorithm, unit tests, and code for running experiments. It additionally contains a fork where certain bugs in the reward_machines (Icarte et al. 2018) repository have been corrected. The corrections are specification of a deterministic reward machine, specifically for CraftWorld. For specific details about the corrections, please refer to the Appendix of the paper.

The main README for REMAP is found at remap/README.

To cite this work:

@misc{hsiung2023remap,
        title={Automata Learning from Preference and Equivalence Queries}, 
        author={Eric Hsiung and Joydeep Biswas and Swarat Chaudhuri},
        year={2023},
        eprint={2308.09301},
        archivePrefix={arXiv},
        primaryClass={cs.LG}
      }