Misspecification in inverse reinforcement learning

J Skalse, A Abate - Proceedings of the AAAI Conference on Artificial …, 2023 - ojs.aaai.org
Abstract The aim of Inverse Reinforcement Learning (IRL) is to infer a reward function R from
a policy pi. To do this, we need a model of how pi relates to R. In the current literature, the …
被引用次数：31 相关文章

Misspecification in inverse reinforcement learning

J Skalse, A Abate - Proceedings of the Thirty-Seventh AAAI Conference …, 2023 - dl.acm.org
The aim of Inverse Reinforcement Learning (IRL) is to infer a reward function R from a policy
π. To do this, we need a model of how π relates to R. In the current literature, the most …

Misspecification in Inverse Reinforcement Learning

J Skalse, A Abate - arXiv preprint arXiv:2212.03201, 2022 - arxiv.org
The aim of Inverse Reinforcement Learning (IRL) is to infer a reward function $ R $ from a
policy $\pi $. To do this, we need a model of how $\pi $ relates to $ R $. In the current …

Misspecification in Inverse Reinforcement Learning

J Skalse, A Abate - arXiv e-prints, 2022 - ui.adsabs.harvard.edu
Abstract The aim of Inverse Reinforcement Learning (IRL) is to infer a reward function $ R $
from a policy $\pi $. To do this, we need a model of how $\pi $ relates to $ R $. In the current …

Misspecification in Inverse Reinforcement Learning

JMV Skalse, A Abate - NeurIPS ML Safety Workshop - openreview.net
The aim of Inverse Reinforcement Learning (IRL) is to infer a reward function $ R $ from a
policy $\pi $. To do this, we need a model of how $\pi $ relates to $ R $. In the current …