AtlasD: Automatic Local Symmetry Discovery
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916792346083328 |
|---|---|
| author | Bhat, Manu Park, Jonghyun Yang, Jianke Dehmamy, Nima Walters, Robin Yu, Rose |
| author_facet | Bhat, Manu Park, Jonghyun Yang, Jianke Dehmamy, Nima Walters, Robin Yu, Rose |
| contents | Existing symmetry discovery methods predominantly focus on global transformations across the entire system or space, but they fail to consider the symmetries in local neighborhoods. This may result in the reported symmetry group being a misrepresentation of the true symmetry. In this paper, we formalize the notion of local symmetry as atlas equivariance. Our proposed pipeline, automatic local symmetry discovery (AtlasD), recovers the local symmetries of a function by training local predictor networks and then learning a Lie group basis to which the predictors are equivariant. We demonstrate AtlasD is capable of discovering local symmetry groups with multiple connected components in top-quark tagging and partial differential equation experiments. The discovered local symmetry is shown to be a useful inductive bias that improves the performance of downstream tasks in climate segmentation and vision tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_10777 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AtlasD: Automatic Local Symmetry Discovery Bhat, Manu Park, Jonghyun Yang, Jianke Dehmamy, Nima Walters, Robin Yu, Rose Machine Learning Existing symmetry discovery methods predominantly focus on global transformations across the entire system or space, but they fail to consider the symmetries in local neighborhoods. This may result in the reported symmetry group being a misrepresentation of the true symmetry. In this paper, we formalize the notion of local symmetry as atlas equivariance. Our proposed pipeline, automatic local symmetry discovery (AtlasD), recovers the local symmetries of a function by training local predictor networks and then learning a Lie group basis to which the predictors are equivariant. We demonstrate AtlasD is capable of discovering local symmetry groups with multiple connected components in top-quark tagging and partial differential equation experiments. The discovered local symmetry is shown to be a useful inductive bias that improves the performance of downstream tasks in climate segmentation and vision tasks. |
| title | AtlasD: Automatic Local Symmetry Discovery |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2504.10777 |