General and Efficient Visual Goal-Conditioned Reinforcement Learning using Object-Agnostic Masks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912634372096000 |
|---|---|
| author | Shahriar, Fahim Wang, Cheryl Azimi, Alireza Vasan, Gautham Elanwar, Hany Hamed Mahmood, A. Rupam Bellinger, Colin |
| author_facet | Shahriar, Fahim Wang, Cheryl Azimi, Alireza Vasan, Gautham Elanwar, Hany Hamed Mahmood, A. Rupam Bellinger, Colin |
| contents | Goal-conditioned reinforcement learning (GCRL) allows agents to learn diverse objectives using a unified policy. The success of GCRL, however, is contingent on the choice of goal representation. In this work, we propose a mask-based goal representation system that provides object-agnostic visual cues to the agent, enabling efficient learning and superior generalization. In contrast, existing goal representation methods, such as target state images, 3D coordinates, and one-hot vectors, face issues of poor generalization to unseen objects, slow convergence, and the need for special cameras. Masks can be processed to generate dense rewards without requiring error-prone distance calculations. Learning with ground truth masks in simulation, we achieved 99.9% reaching accuracy on training and unseen test objects. Our proposed method can be utilized to perform pick-up tasks with high accuracy, without using any positional information of the target. Moreover, we demonstrate learning from scratch and sim-to-real transfer applications using two different physical robots, utilizing pretrained open vocabulary object detection models for mask generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_06277 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | General and Efficient Visual Goal-Conditioned Reinforcement Learning using Object-Agnostic Masks Shahriar, Fahim Wang, Cheryl Azimi, Alireza Vasan, Gautham Elanwar, Hany Hamed Mahmood, A. Rupam Bellinger, Colin Computer Vision and Pattern Recognition Machine Learning Goal-conditioned reinforcement learning (GCRL) allows agents to learn diverse objectives using a unified policy. The success of GCRL, however, is contingent on the choice of goal representation. In this work, we propose a mask-based goal representation system that provides object-agnostic visual cues to the agent, enabling efficient learning and superior generalization. In contrast, existing goal representation methods, such as target state images, 3D coordinates, and one-hot vectors, face issues of poor generalization to unseen objects, slow convergence, and the need for special cameras. Masks can be processed to generate dense rewards without requiring error-prone distance calculations. Learning with ground truth masks in simulation, we achieved 99.9% reaching accuracy on training and unseen test objects. Our proposed method can be utilized to perform pick-up tasks with high accuracy, without using any positional information of the target. Moreover, we demonstrate learning from scratch and sim-to-real transfer applications using two different physical robots, utilizing pretrained open vocabulary object detection models for mask generation. |
| title | General and Efficient Visual Goal-Conditioned Reinforcement Learning using Object-Agnostic Masks |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2510.06277 |