Emergent Neural Automaton Policies: Learning Symbolic Structure from Visuomotor Trajectories
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_ | 1866914476468469760 |
|---|---|
| author | Pan, Yiyuan Luo, Xusheng Hu, Hanjiang Yu, Peiqi Liu, Changliu |
| author_facet | Pan, Yiyuan Luo, Xusheng Hu, Hanjiang Yu, Peiqi Liu, Changliu |
| contents | Scaling robot learning to long-horizon tasks remains a formidable challenge. While end-to-end policies often lack the structural priors needed for effective long-term reasoning, traditional neuro-symbolic methods rely heavily on hand-crafted symbolic priors. To address the issue, we introduce ENAP (Emergent Neural Automaton Policy), a framework that allows a bi-level neuro-symbolic policy adaptively emerge from visuomotor demonstrations. Specifically, we first employ adaptive clustering and an extension of the L* algorithm to infer a Mealy state machine from visuomotor data, which serves as an interpretable high-level planner capturing latent task modes. Then, this discrete structure guides a low-level reactive residual network to learn precise continuous control via behavior cloning (BC). By explicitly modeling the task structure with discrete transitions and continuous residuals, ENAP achieves high sample efficiency and interpretability without requiring task-specific labels. Extensive experiments on complex manipulation and long-horizon tasks demonstrate that ENAP outperforms state-of-the-art (SoTA) end-to-end VLA policies by up to 27% in low-data regimes, while offering a structured representation of robotic intent (Fig. 1). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_25903 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Emergent Neural Automaton Policies: Learning Symbolic Structure from Visuomotor Trajectories Pan, Yiyuan Luo, Xusheng Hu, Hanjiang Yu, Peiqi Liu, Changliu Robotics Scaling robot learning to long-horizon tasks remains a formidable challenge. While end-to-end policies often lack the structural priors needed for effective long-term reasoning, traditional neuro-symbolic methods rely heavily on hand-crafted symbolic priors. To address the issue, we introduce ENAP (Emergent Neural Automaton Policy), a framework that allows a bi-level neuro-symbolic policy adaptively emerge from visuomotor demonstrations. Specifically, we first employ adaptive clustering and an extension of the L* algorithm to infer a Mealy state machine from visuomotor data, which serves as an interpretable high-level planner capturing latent task modes. Then, this discrete structure guides a low-level reactive residual network to learn precise continuous control via behavior cloning (BC). By explicitly modeling the task structure with discrete transitions and continuous residuals, ENAP achieves high sample efficiency and interpretability without requiring task-specific labels. Extensive experiments on complex manipulation and long-horizon tasks demonstrate that ENAP outperforms state-of-the-art (SoTA) end-to-end VLA policies by up to 27% in low-data regimes, while offering a structured representation of robotic intent (Fig. 1). |
| title | Emergent Neural Automaton Policies: Learning Symbolic Structure from Visuomotor Trajectories |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.25903 |