Sustainable Transfer Learning for Adaptive Robot Skills
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915923854622720 |
|---|---|
| author | Abuibaid, Khalil Hegiste, Vinit Gafur, Nigora Wagner, Achim Ruskowski, Martin |
| author_facet | Abuibaid, Khalil Hegiste, Vinit Gafur, Nigora Wagner, Achim Ruskowski, Martin |
| contents | Learning robot skills from scratch is often time-consuming, while reusing data promotes sustainability and improves sample efficiency. This study investigates policy transfer across different robotic platforms, focusing on peg-in-hole task using reinforcement learning (RL). Policy training is carried out on two different robots. Their policies are transferred and evaluated for zero-shot, fine-tuning, and training from scratch. Results indicate that zero-shot transfer leads to lower success rates and relatively longer task execution times, while fine-tuning significantly improves performance with fewer training time-steps. These findings highlight that policy transfer with adaptation techniques improves sample efficiency and generalization, reducing the need for extensive retraining and supporting sustainable robotic learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_06943 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Sustainable Transfer Learning for Adaptive Robot Skills Abuibaid, Khalil Hegiste, Vinit Gafur, Nigora Wagner, Achim Ruskowski, Martin Robotics Learning robot skills from scratch is often time-consuming, while reusing data promotes sustainability and improves sample efficiency. This study investigates policy transfer across different robotic platforms, focusing on peg-in-hole task using reinforcement learning (RL). Policy training is carried out on two different robots. Their policies are transferred and evaluated for zero-shot, fine-tuning, and training from scratch. Results indicate that zero-shot transfer leads to lower success rates and relatively longer task execution times, while fine-tuning significantly improves performance with fewer training time-steps. These findings highlight that policy transfer with adaptation techniques improves sample efficiency and generalization, reducing the need for extensive retraining and supporting sustainable robotic learning. |
| title | Sustainable Transfer Learning for Adaptive Robot Skills |
| topic | Robotics |
| url | https://arxiv.org/abs/2604.06943 |