Robust Inverse Constrained Reinforcement Learning under Model Misspecification

Sheng Xu, Guiliang Liu
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:55162-55185, 2024.

Abstract

To solve safety-critical decision-making problems, Inverse Constrained Reinforcement Learning (ICRL) infers constraints from expert demonstrations and seeks to imitate expert preference by utilizing these constraints. While prior ICRL research commonly overlooks the discrepancy between the training and deploying environments, we demonstrate that such a discrepancy can significantly compromise the reliability of the inferred constraints and thus induce unsafe movements. Motivated by this finding, we propose the Robust Constraint Inference (RCI) problem and an Adaptively Robust ICRL (AR-ICRL) algorithm to solve RCI efficiently. Specifically, we model the impact of misspecified dynamics with an opponent policy and learn a robust policy to facilitate safe control in a Markov Game. Subsequently, we adjust our constraint model to align the learned policies to expert demonstrations, accommodating both soft and hard optimality in our behavioral models. Empirical results demonstrate the significance of robust constraints and the effectiveness of the proposed AR-ICRL algorithm under continuous and discrete domains. The code is available at https://github.com/Jasonxu1225/AR-ICRL.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-xu24r, title = {Robust Inverse Constrained Reinforcement Learning under Model Misspecification}, author = {Xu, Sheng and Liu, Guiliang}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {55162--55185}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/xu24r/xu24r.pdf}, url = {https://proceedings.mlr.press/v235/xu24r.html}, abstract = {To solve safety-critical decision-making problems, Inverse Constrained Reinforcement Learning (ICRL) infers constraints from expert demonstrations and seeks to imitate expert preference by utilizing these constraints. While prior ICRL research commonly overlooks the discrepancy between the training and deploying environments, we demonstrate that such a discrepancy can significantly compromise the reliability of the inferred constraints and thus induce unsafe movements. Motivated by this finding, we propose the Robust Constraint Inference (RCI) problem and an Adaptively Robust ICRL (AR-ICRL) algorithm to solve RCI efficiently. Specifically, we model the impact of misspecified dynamics with an opponent policy and learn a robust policy to facilitate safe control in a Markov Game. Subsequently, we adjust our constraint model to align the learned policies to expert demonstrations, accommodating both soft and hard optimality in our behavioral models. Empirical results demonstrate the significance of robust constraints and the effectiveness of the proposed AR-ICRL algorithm under continuous and discrete domains. The code is available at https://github.com/Jasonxu1225/AR-ICRL.} }
Endnote
%0 Conference Paper %T Robust Inverse Constrained Reinforcement Learning under Model Misspecification %A Sheng Xu %A Guiliang Liu %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-xu24r %I PMLR %P 55162--55185 %U https://proceedings.mlr.press/v235/xu24r.html %V 235 %X To solve safety-critical decision-making problems, Inverse Constrained Reinforcement Learning (ICRL) infers constraints from expert demonstrations and seeks to imitate expert preference by utilizing these constraints. While prior ICRL research commonly overlooks the discrepancy between the training and deploying environments, we demonstrate that such a discrepancy can significantly compromise the reliability of the inferred constraints and thus induce unsafe movements. Motivated by this finding, we propose the Robust Constraint Inference (RCI) problem and an Adaptively Robust ICRL (AR-ICRL) algorithm to solve RCI efficiently. Specifically, we model the impact of misspecified dynamics with an opponent policy and learn a robust policy to facilitate safe control in a Markov Game. Subsequently, we adjust our constraint model to align the learned policies to expert demonstrations, accommodating both soft and hard optimality in our behavioral models. Empirical results demonstrate the significance of robust constraints and the effectiveness of the proposed AR-ICRL algorithm under continuous and discrete domains. The code is available at https://github.com/Jasonxu1225/AR-ICRL.
APA
Xu, S. & Liu, G.. (2024). Robust Inverse Constrained Reinforcement Learning under Model Misspecification. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:55162-55185 Available from https://proceedings.mlr.press/v235/xu24r.html.

Related Material