Task Parameter Extrapolation via Learning Inverse Tasks from Forward Demonstrations
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912953931923456 |
|---|---|
| author | Bahar, Serdar Dogangun, Fatih Saveriano, Matteo Nagai, Yukie Ugur, Emre |
| author_facet | Bahar, Serdar Dogangun, Fatih Saveriano, Matteo Nagai, Yukie Ugur, Emre |
| contents | Generalizing skill policies to novel conditions remains a key challenge in robot learning. Imitation learning methods, while data-efficient, are largely confined to the training region and consistently fail on input data outside it, leading to unpredictable policy failures. Alternatively, transfer learning approaches offer methods for trajectory generation robust to both changes in environment or tasks, but they remain data-hungry and lack accuracy in zero-shot generalization. We address these challenges by framing the problem in the context of task inversion learning and proposing a novel joint learning approach to achieve accurate and efficient knowledge transfer. Our method constructs a common representation of the forward and inverse tasks, and leverages auxiliary forward demonstrations from novel configurations to successfully execute the corresponding inverse tasks, without any direct supervision. We show the extrapolation capabilities of our framework via ablation studies and experiments in simulated and real-world environments that require complex manipulation skills with a diverse set of objects and tools, where we outperform diffusion-based alternatives. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_05576 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Task Parameter Extrapolation via Learning Inverse Tasks from Forward Demonstrations Bahar, Serdar Dogangun, Fatih Saveriano, Matteo Nagai, Yukie Ugur, Emre Robotics Generalizing skill policies to novel conditions remains a key challenge in robot learning. Imitation learning methods, while data-efficient, are largely confined to the training region and consistently fail on input data outside it, leading to unpredictable policy failures. Alternatively, transfer learning approaches offer methods for trajectory generation robust to both changes in environment or tasks, but they remain data-hungry and lack accuracy in zero-shot generalization. We address these challenges by framing the problem in the context of task inversion learning and proposing a novel joint learning approach to achieve accurate and efficient knowledge transfer. Our method constructs a common representation of the forward and inverse tasks, and leverages auxiliary forward demonstrations from novel configurations to successfully execute the corresponding inverse tasks, without any direct supervision. We show the extrapolation capabilities of our framework via ablation studies and experiments in simulated and real-world environments that require complex manipulation skills with a diverse set of objects and tools, where we outperform diffusion-based alternatives. |
| title | Task Parameter Extrapolation via Learning Inverse Tasks from Forward Demonstrations |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.05576 |