Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2506.05418 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918046576148480 |
|---|---|
| author | Kim, Kyungsoo Ha, Jeongsoo Kim, Yusung |
| author_facet | Kim, Kyungsoo Ha, Jeongsoo Kim, Yusung |
| contents | Vision-based reinforcement learning requires efficient and robust representations of image-based observations, especially when the images contain distracting (task-irrelevant) elements such as shadows, clouds, and light. It becomes more important if those distractions are not exposed during training. We design a Self-Predictive Dynamics (SPD) method to extract task-relevant features efficiently, even in unseen observations after training. SPD uses weak and strong augmentations in parallel, and learns representations by predicting inverse and forward transitions across the two-way augmented versions. In a set of MuJoCo visual control tasks and an autonomous driving task (CARLA), SPD outperforms previous studies in complex observations, and significantly improves the generalization performance for unseen observations. Our code is available at https://github.com/unigary/SPD. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_05418 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Self-Predictive Dynamics for Generalization of Vision-based Reinforcement Learning Kim, Kyungsoo Ha, Jeongsoo Kim, Yusung Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Vision-based reinforcement learning requires efficient and robust representations of image-based observations, especially when the images contain distracting (task-irrelevant) elements such as shadows, clouds, and light. It becomes more important if those distractions are not exposed during training. We design a Self-Predictive Dynamics (SPD) method to extract task-relevant features efficiently, even in unseen observations after training. SPD uses weak and strong augmentations in parallel, and learns representations by predicting inverse and forward transitions across the two-way augmented versions. In a set of MuJoCo visual control tasks and an autonomous driving task (CARLA), SPD outperforms previous studies in complex observations, and significantly improves the generalization performance for unseen observations. Our code is available at https://github.com/unigary/SPD. |
| title | Self-Predictive Dynamics for Generalization of Vision-based Reinforcement Learning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2506.05418 |