Zero-Shot Offline Imitation Learning via Optimal Transport

Thomas Rupf, Marco Bagatella, Nico Gürtler, Jonas Frey, Georg Martius
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:52345-52381, 2025.

Abstract

Zero-shot imitation learning algorithms hold the promise of reproducing unseen behavior from as little as a single demonstration at test time. Existing practical approaches view the expert demonstration as a sequence of goals, enabling imitation with a high-level goal selector, and a low-level goal-conditioned policy. However, this framework can suffer from myopic behavior: the agent’s immediate actions towards achieving individual goals may undermine long-term objectives. We introduce a novel method that mitigates this issue by directly optimizing the occupancy matching objective that is intrinsic to imitation learning. We propose to lift a goal-conditioned value function to a distance between occupancies, which are in turn approximated via a learned world model. The resulting method can learn from offline, suboptimal data, and is capable of non-myopic, zero-shot imitation, as we demonstrate in complex, continuous benchmarks. The code is available at https://github.com/martius-lab/zilot.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-rupf25a, title = {Zero-Shot Offline Imitation Learning via Optimal Transport}, author = {Rupf, Thomas and Bagatella, Marco and G\"{u}rtler, Nico and Frey, Jonas and Martius, Georg}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {52345--52381}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/rupf25a/rupf25a.pdf}, url = {https://proceedings.mlr.press/v267/rupf25a.html}, abstract = {Zero-shot imitation learning algorithms hold the promise of reproducing unseen behavior from as little as a single demonstration at test time. Existing practical approaches view the expert demonstration as a sequence of goals, enabling imitation with a high-level goal selector, and a low-level goal-conditioned policy. However, this framework can suffer from myopic behavior: the agent’s immediate actions towards achieving individual goals may undermine long-term objectives. We introduce a novel method that mitigates this issue by directly optimizing the occupancy matching objective that is intrinsic to imitation learning. We propose to lift a goal-conditioned value function to a distance between occupancies, which are in turn approximated via a learned world model. The resulting method can learn from offline, suboptimal data, and is capable of non-myopic, zero-shot imitation, as we demonstrate in complex, continuous benchmarks. The code is available at https://github.com/martius-lab/zilot.} }
Endnote
%0 Conference Paper %T Zero-Shot Offline Imitation Learning via Optimal Transport %A Thomas Rupf %A Marco Bagatella %A Nico Gürtler %A Jonas Frey %A Georg Martius %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-rupf25a %I PMLR %P 52345--52381 %U https://proceedings.mlr.press/v267/rupf25a.html %V 267 %X Zero-shot imitation learning algorithms hold the promise of reproducing unseen behavior from as little as a single demonstration at test time. Existing practical approaches view the expert demonstration as a sequence of goals, enabling imitation with a high-level goal selector, and a low-level goal-conditioned policy. However, this framework can suffer from myopic behavior: the agent’s immediate actions towards achieving individual goals may undermine long-term objectives. We introduce a novel method that mitigates this issue by directly optimizing the occupancy matching objective that is intrinsic to imitation learning. We propose to lift a goal-conditioned value function to a distance between occupancies, which are in turn approximated via a learned world model. The resulting method can learn from offline, suboptimal data, and is capable of non-myopic, zero-shot imitation, as we demonstrate in complex, continuous benchmarks. The code is available at https://github.com/martius-lab/zilot.
APA
Rupf, T., Bagatella, M., Gürtler, N., Frey, J. & Martius, G.. (2025). Zero-Shot Offline Imitation Learning via Optimal Transport. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:52345-52381 Available from https://proceedings.mlr.press/v267/rupf25a.html.

Related Material