Mean Flow Policy with Instantaneous Velocity Constraint for One-step Action Generation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918377567551488 |
|---|---|
| author | Zhan, Guojian Tao, Letian Wang, Pengcheng Wang, Yixiao Li, Yiheng Chen, Yuxin Li, Hongyang Tomizuka, Masayoshi Li, Shengbo Eben |
| author_facet | Zhan, Guojian Tao, Letian Wang, Pengcheng Wang, Yixiao Li, Yiheng Chen, Yuxin Li, Hongyang Tomizuka, Masayoshi Li, Shengbo Eben |
| contents | Learning expressive and efficient policy functions is a promising direction in reinforcement learning (RL). While flow-based policies have recently proven effective in modeling complex action distributions with a fast deterministic sampling process, they still face a trade-off between expressiveness and computational burden, which is typically controlled by the number of flow steps. In this work, we propose mean velocity policy (MVP), a new generative policy function that models the mean velocity field to achieve the fastest one-step action generation. To ensure its high expressiveness, an instantaneous velocity constraint (IVC) is introduced on the mean velocity field during training. We theoretically prove that this design explicitly serves as a crucial boundary condition, thereby improving learning accuracy and enhancing policy expressiveness. Empirically, our MVP achieves state-of-the-art success rates across several challenging robotic manipulation tasks from Robomimic and OGBench. It also delivers substantial improvements in training and inference speed over existing flow-based policy baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_13810 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Mean Flow Policy with Instantaneous Velocity Constraint for One-step Action Generation Zhan, Guojian Tao, Letian Wang, Pengcheng Wang, Yixiao Li, Yiheng Chen, Yuxin Li, Hongyang Tomizuka, Masayoshi Li, Shengbo Eben Machine Learning Artificial Intelligence Learning expressive and efficient policy functions is a promising direction in reinforcement learning (RL). While flow-based policies have recently proven effective in modeling complex action distributions with a fast deterministic sampling process, they still face a trade-off between expressiveness and computational burden, which is typically controlled by the number of flow steps. In this work, we propose mean velocity policy (MVP), a new generative policy function that models the mean velocity field to achieve the fastest one-step action generation. To ensure its high expressiveness, an instantaneous velocity constraint (IVC) is introduced on the mean velocity field during training. We theoretically prove that this design explicitly serves as a crucial boundary condition, thereby improving learning accuracy and enhancing policy expressiveness. Empirically, our MVP achieves state-of-the-art success rates across several challenging robotic manipulation tasks from Robomimic and OGBench. It also delivers substantial improvements in training and inference speed over existing flow-based policy baselines. |
| title | Mean Flow Policy with Instantaneous Velocity Constraint for One-step Action Generation |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2602.13810 |