E(n) Equivariant Normalizing Flows
This paper introduces a generative model equivariant to Euclidean symmetries: E(n) Equivariant Normalizing Flows (E-NFs). To construct E-NFs, we take the discriminative E(n) graph neural networks and integrate them as a differential equation to obtain an invertible equivariant function: a continuous...
| Main Authors: | , , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
Curran Associates
2022
|
| _version_ | 1826309445921013760 |
|---|---|
| author | Satorras, VG Hoogeboom, E Fuchs, FB Posner, I Welling, M |
| author_facet | Satorras, VG Hoogeboom, E Fuchs, FB Posner, I Welling, M |
| author_sort | Satorras, VG |
| collection | OXFORD |
| description | This paper introduces a generative model equivariant to Euclidean symmetries: E(n) Equivariant Normalizing Flows (E-NFs). To construct E-NFs, we take the discriminative E(n) graph neural networks and integrate them as a differential equation to obtain an invertible equivariant function: a continuous-time normalizing flow. We demonstrate that E-NFs considerably outperform baselines and existing methods from the literature on particle systems such as DW4 and LJ13, and on molecules from QM9 in terms of log-likelihood. To the best of our knowledge, this is the first flow that jointly generates molecule features and positions in 3D. |
| first_indexed | 2024-03-07T07:35:49Z |
| format | Conference item |
| id | oxford-uuid:04a812fe-72d4-4992-94a9-561aa037ba80 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:35:49Z |
| publishDate | 2022 |
| publisher | Curran Associates |
| record_format | dspace |
| spelling | oxford-uuid:04a812fe-72d4-4992-94a9-561aa037ba802023-03-14T14:37:04ZE(n) Equivariant Normalizing FlowsConference itemhttp://purl.org/coar/resource_type/c_5794uuid:04a812fe-72d4-4992-94a9-561aa037ba80EnglishSymplectic ElementsCurran Associates2022Satorras, VGHoogeboom, EFuchs, FBPosner, IWelling, MThis paper introduces a generative model equivariant to Euclidean symmetries: E(n) Equivariant Normalizing Flows (E-NFs). To construct E-NFs, we take the discriminative E(n) graph neural networks and integrate them as a differential equation to obtain an invertible equivariant function: a continuous-time normalizing flow. We demonstrate that E-NFs considerably outperform baselines and existing methods from the literature on particle systems such as DW4 and LJ13, and on molecules from QM9 in terms of log-likelihood. To the best of our knowledge, this is the first flow that jointly generates molecule features and positions in 3D. |
| spellingShingle | Satorras, VG Hoogeboom, E Fuchs, FB Posner, I Welling, M E(n) Equivariant Normalizing Flows |
| title | E(n) Equivariant Normalizing Flows |
| title_full | E(n) Equivariant Normalizing Flows |
| title_fullStr | E(n) Equivariant Normalizing Flows |
| title_full_unstemmed | E(n) Equivariant Normalizing Flows |
| title_short | E(n) Equivariant Normalizing Flows |
| title_sort | e n equivariant normalizing flows |
| work_keys_str_mv | AT satorrasvg enequivariantnormalizingflows AT hoogeboome enequivariantnormalizingflows AT fuchsfb enequivariantnormalizingflows AT posneri enequivariantnormalizingflows AT wellingm enequivariantnormalizingflows |