[1]
Booth, S. et al. 2023. The Perils of Trial-and-Error Reward Design: Misdesign through Overfitting and Invalid Task Specifications. Proceedings of the AAAI Conference on Artificial Intelligence. 37, 5 (Jun. 2023), 5920–5929. DOI:https://doi.org/10.1609/aaai.v37i5.25733.