Relaxed Equivariant Graph Neural Networks
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912150369337344 |
|---|---|
| author | Hofgard, Elyssa Wang, Rui Walters, Robin Smidt, Tess |
| author_facet | Hofgard, Elyssa Wang, Rui Walters, Robin Smidt, Tess |
| contents | 3D Euclidean symmetry equivariant neural networks have demonstrated notable success in modeling complex physical systems. We introduce a framework for relaxed $E(3)$ graph equivariant neural networks that can learn and represent symmetry breaking within continuous groups. Building on the existing e3nn framework, we propose the use of relaxed weights to allow for controlled symmetry breaking. We show empirically that these relaxed weights learn the correct amount of symmetry breaking. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_20471 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Relaxed Equivariant Graph Neural Networks Hofgard, Elyssa Wang, Rui Walters, Robin Smidt, Tess Machine Learning 3D Euclidean symmetry equivariant neural networks have demonstrated notable success in modeling complex physical systems. We introduce a framework for relaxed $E(3)$ graph equivariant neural networks that can learn and represent symmetry breaking within continuous groups. Building on the existing e3nn framework, we propose the use of relaxed weights to allow for controlled symmetry breaking. We show empirically that these relaxed weights learn the correct amount of symmetry breaking. |
| title | Relaxed Equivariant Graph Neural Networks |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2407.20471 |