Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2509.26294 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909816731992064 |
|---|---|
| author | Blondé, Lionel Ramos, Joao A. Candido Kalousis, Alexandros |
| author_facet | Blondé, Lionel Ramos, Joao A. Candido Kalousis, Alexandros |
| contents | 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. Code is publicly available at: https://github.com/lionelblonde/ngt-pytorch. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_26294 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Noise-Guided Transport for Imitation Learning Blondé, Lionel Ramos, Joao A. Candido Kalousis, Alexandros Machine Learning Artificial Intelligence 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. Code is publicly available at: https://github.com/lionelblonde/ngt-pytorch. |
| title | Noise-Guided Transport for Imitation Learning |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2509.26294 |