Noise-Guided Transport: Imitation Learning from Random Priors

Lionel Blondé, Joao Candido Ramos, Alexandros Kalousis
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:8618-8643, 2026.

Abstract

We consider imitation learning in the low-data regime, where only a limited number of expert demonstrations are available. In this setting, methods that rely on large-scale pretraining or high-capacity architectures can be difficult to apply, and efficiency with respect to demonstration data becomes critical. We introduce Noise-Guided Transport (NGT), a lightweight off-policy method that casts imitation as an optimal transport problem solved via adversarial training. NGT requires no pretraining or specialized architectures, incorporates uncertainty estimation by design, and is easy to implement and tune. Despite its simplicity, NGT achieves strong performance on challenging continuous control tasks, including high-dimensional Humanoid tasks, under ultra-low data regimes with as few as 20 transitions.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-blonde26a, title = {Noise-Guided Transport: Imitation Learning from Random Priors}, author = {Blond\'{e}, Lionel and Candido Ramos, Joao and Kalousis, Alexandros}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {8618--8643}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/blonde26a/blonde26a.pdf}, url = {https://proceedings.mlr.press/v306/blonde26a.html}, abstract = {We consider imitation learning in the low-data regime, where only a limited number of expert demonstrations are available. In this setting, methods that rely on large-scale pretraining or high-capacity architectures can be difficult to apply, and efficiency with respect to demonstration data becomes critical. We introduce Noise-Guided Transport (NGT), a lightweight off-policy method that casts imitation as an optimal transport problem solved via adversarial training. NGT requires no pretraining or specialized architectures, incorporates uncertainty estimation by design, and is easy to implement and tune. Despite its simplicity, NGT achieves strong performance on challenging continuous control tasks, including high-dimensional Humanoid tasks, under ultra-low data regimes with as few as 20 transitions.} }
Endnote
%0 Conference Paper %T Noise-Guided Transport: Imitation Learning from Random Priors %A Lionel Blondé %A Joao Candido Ramos %A Alexandros Kalousis %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-blonde26a %I PMLR %P 8618--8643 %U https://proceedings.mlr.press/v306/blonde26a.html %V 306 %X We consider imitation learning in the low-data regime, where only a limited number of expert demonstrations are available. In this setting, methods that rely on large-scale pretraining or high-capacity architectures can be difficult to apply, and efficiency with respect to demonstration data becomes critical. We introduce Noise-Guided Transport (NGT), a lightweight off-policy method that casts imitation as an optimal transport problem solved via adversarial training. NGT requires no pretraining or specialized architectures, incorporates uncertainty estimation by design, and is easy to implement and tune. Despite its simplicity, NGT achieves strong performance on challenging continuous control tasks, including high-dimensional Humanoid tasks, under ultra-low data regimes with as few as 20 transitions.
APA
Blondé, L., Candido Ramos, J. & Kalousis, A.. (2026). Noise-Guided Transport: Imitation Learning from Random Priors. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:8618-8643 Available from https://proceedings.mlr.press/v306/blonde26a.html.

Related Material