Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2601.21881 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912859631386624 |
|---|---|
| author | Peng, Jingruo Zhu, Shuze |
| author_facet | Peng, Jingruo Zhu, Shuze |
| contents | Humans can infer accurate mechanical outcomes from only a few observations, a capability known as mechanics intuition. The mechanisms behind such data-efficient learning remain unclear. Here, we propose a reinforcement learning framework in which an agent encodes continuous physical observation parameters into its state and is trained via episodic switching across closely related observations. With merely two or three observations, the agent acquires robust mechanics intuition that generalizes accurately over wide parameter ranges, substantially beyond the training data, as demonstrated on the brachistochrone and a large-deformation elastic plate. We explain this generalization through a unified theoretical view: it emerges when the learned value function enforces Bellman consistency across neighboring task parameters, rendering the Bellman residual stationary with respect to physical variations. This induces a smooth policy that captures a low-dimensional solution manifold underlying the continuum of tasks. Our work establishes episodic switching as a principled route to artificial mechanics intuition and offers a theoretical link to similar generalization abilities in biological learners. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_21881 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Acquiring Human-Like Mechanics Intuition from Scarce Observations via Deep Reinforcement Learning Peng, Jingruo Zhu, Shuze Computational Physics Humans can infer accurate mechanical outcomes from only a few observations, a capability known as mechanics intuition. The mechanisms behind such data-efficient learning remain unclear. Here, we propose a reinforcement learning framework in which an agent encodes continuous physical observation parameters into its state and is trained via episodic switching across closely related observations. With merely two or three observations, the agent acquires robust mechanics intuition that generalizes accurately over wide parameter ranges, substantially beyond the training data, as demonstrated on the brachistochrone and a large-deformation elastic plate. We explain this generalization through a unified theoretical view: it emerges when the learned value function enforces Bellman consistency across neighboring task parameters, rendering the Bellman residual stationary with respect to physical variations. This induces a smooth policy that captures a low-dimensional solution manifold underlying the continuum of tasks. Our work establishes episodic switching as a principled route to artificial mechanics intuition and offers a theoretical link to similar generalization abilities in biological learners. |
| title | Acquiring Human-Like Mechanics Intuition from Scarce Observations via Deep Reinforcement Learning |
| topic | Computational Physics |
| url | https://arxiv.org/abs/2601.21881 |