Adversarial Inverse Optimal Control for General Imitation Learning Losses and Embodiment Transfer

Xiangli Chen UIC, Mathew Monfort, Brian Ziebart, Peter Carr
Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, PMLR R14:326-335, 2016.

Abstract

We develop a general framework for inverse optimal control that distinguishes between rationalizing demonstrated behavior and imitating inductively inferred behavior. This enables learning for more general imitative evaluation measures and differences between the capabilities of the demonstrator and those of the learner (i.e., differences in embodiment). Our formulation takes the form of a zero-sum game between a predictor attempting to minimize an imitative loss measure, and an adversary attempting to maximize the loss by approximating the demonstrated examples in limited ways. We establish the consistency and generalization guarantees of this approach and il- lustrate its benefits on real and synthetic imitation learning tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR14-uic16a, title = {Adversarial Inverse Optimal Control for General Imitation Learning Losses and Embodiment Transfer}, author = {UIC, Xiangli Chen and Monfort, Mathew and Ziebart, Brian and Carr, Peter}, booktitle = {Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence}, pages = {326--335}, year = {2016}, editor = {Ihler, Alexander and Janzing, Dominik}, volume = {R14}, series = {Proceedings of Machine Learning Research}, month = {25--29 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r14/main/assets/uic16a/uic16a.pdf}, url = {https://proceedings.mlr.press/r14/uic16a.html}, abstract = {We develop a general framework for inverse optimal control that distinguishes between rationalizing demonstrated behavior and imitating inductively inferred behavior. This enables learning for more general imitative evaluation measures and differences between the capabilities of the demonstrator and those of the learner (i.e., differences in embodiment). Our formulation takes the form of a zero-sum game between a predictor attempting to minimize an imitative loss measure, and an adversary attempting to maximize the loss by approximating the demonstrated examples in limited ways. We establish the consistency and generalization guarantees of this approach and il- lustrate its benefits on real and synthetic imitation learning tasks.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Adversarial Inverse Optimal Control for General Imitation Learning Losses and Embodiment Transfer %A Xiangli Chen UIC %A Mathew Monfort %A Brian Ziebart %A Peter Carr %B Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2016 %E Alexander Ihler %E Dominik Janzing %F pmlr-vR14-uic16a %I PMLR %P 326--335 %U https://proceedings.mlr.press/r14/uic16a.html %V R14 %X We develop a general framework for inverse optimal control that distinguishes between rationalizing demonstrated behavior and imitating inductively inferred behavior. This enables learning for more general imitative evaluation measures and differences between the capabilities of the demonstrator and those of the learner (i.e., differences in embodiment). Our formulation takes the form of a zero-sum game between a predictor attempting to minimize an imitative loss measure, and an adversary attempting to maximize the loss by approximating the demonstrated examples in limited ways. We establish the consistency and generalization guarantees of this approach and il- lustrate its benefits on real and synthetic imitation learning tasks. %Z Reissued by PMLR on 04 October 2026.
APA
UIC, X.C., Monfort, M., Ziebart, B. & Carr, P.. (2016). Adversarial Inverse Optimal Control for General Imitation Learning Losses and Embodiment Transfer. Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R14:326-335 Available from https://proceedings.mlr.press/r14/uic16a.html. Reissued by PMLR on 04 October 2026.

Related Material