Booth, S., Knox, W. B., Shah, J., Niekum, S., Stone, P., & Allievi, A. (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), 5920-5929. https://doi.org/10.1609/aaai.v37i5.25733