Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2502.01316 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908550656163840 |
|---|---|
| author | Wang, Zeyu Li, Yao-Hui Li, Xin Zang, Hongyu Laroche, Romain Islam, Riashat |
| author_facet | Wang, Zeyu Li, Yao-Hui Li, Xin Zang, Hongyu Laroche, Romain Islam, Riashat |
| contents | Multi-View Reinforcement Learning (MVRL) seeks to provide agents with multi-view observations, enabling them to perceive environment with greater effectiveness and precision. Recent advancements in MVRL focus on extracting latent representations from multiview observations and leveraging them in control tasks. However, it is not straightforward to learn compact and task-relevant representations, particularly in the presence of redundancy, distracting information, or missing views. In this paper, we propose Multi-view Fusion State for Control (MFSC), firstly incorporating bisimulation metric learning into MVRL to learn task-relevant representations. Furthermore, we propose a multiview-based mask and latent reconstruction auxiliary task that exploits shared information across views and improves MFSC's robustness in missing views by introducing a mask token. Extensive experimental results demonstrate that our method outperforms existing approaches in MVRL tasks. Even in more realistic scenarios with interference or missing views, MFSC consistently maintains high performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_01316 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning Fused State Representations for Control from Multi-View Observations Wang, Zeyu Li, Yao-Hui Li, Xin Zang, Hongyu Laroche, Romain Islam, Riashat Machine Learning Artificial Intelligence Multi-View Reinforcement Learning (MVRL) seeks to provide agents with multi-view observations, enabling them to perceive environment with greater effectiveness and precision. Recent advancements in MVRL focus on extracting latent representations from multiview observations and leveraging them in control tasks. However, it is not straightforward to learn compact and task-relevant representations, particularly in the presence of redundancy, distracting information, or missing views. In this paper, we propose Multi-view Fusion State for Control (MFSC), firstly incorporating bisimulation metric learning into MVRL to learn task-relevant representations. Furthermore, we propose a multiview-based mask and latent reconstruction auxiliary task that exploits shared information across views and improves MFSC's robustness in missing views by introducing a mask token. Extensive experimental results demonstrate that our method outperforms existing approaches in MVRL tasks. Even in more realistic scenarios with interference or missing views, MFSC consistently maintains high performance. |
| title | Learning Fused State Representations for Control from Multi-View Observations |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2502.01316 |