Misalignment from Treating Means as Ends
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (44) · 37719-37727
DOI: 10.1609/aaai.v40i44.41107 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Henrik Marklund und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Reward functions, learned or manually specified, are rarely perfect. Instead of accurately expressing human goals, these reward functions are often distorted by human beliefs about how best to achieve those goals. Specifically, these reward functions often express a combination of the human's terminal goals — those which are ends in themselves — and the human's instrumental goals — those which are means to an end. We formulate a simple example in which even slight conflation of instrumental and terminal goals results in severe misalignment: optimizing the misspecified reward function r̂ results in poor performance when measured by the true reward function r. This example distills the essential properties of environments that make reinforcement learning highly sensitive to conflation of instrumental and terminal goals. We discuss how this issue can arise with a common approach to reward learning and how it can manifest in real environments.
Zitieren
APA
Marklund, H., Infanger, A., & Roy, B. V. (2026). Misalignment from Treating Means as Ends. Proceedings of the AAAI Conference on Artificial Intelligence, 40(44), 37719-37727. https://doi.org/10.1609/aaai.v40i44.41107
Vancouver
Marklund H, Infanger A, Roy BV. Misalignment from Treating Means as Ends. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(44):37719-37727. doi:10.1609/aaai.v40i44.41107
BibTeX
@article{Marklund2026,
author = {Marklund, Henrik and Infanger, Alex and Roy, Benjamin Van},
title = {Misalignment from Treating Means as Ends},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {44},
pages = {37719--37727},
doi = {10.1609/aaai.v40i44.41107}
}
EUVIMED