Energy Guided Geometric Flow Matching
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918205091479552 |
|---|---|
| author | Zweig, Aaron Zhang, Mingxuan Azizi, Elham Knowles, David |
| author_facet | Zweig, Aaron Zhang, Mingxuan Azizi, Elham Knowles, David |
| contents | A useful inductive bias for temporal data is that trajectories should stay close to the data manifold. Traditional flow matching relies on straight conditional paths, and flow matching methods which learn geodesics rely on RBF kernels or nearest neighbor graphs that suffer from the curse of dimensionality. We propose to use score matching and annealed energy distillation to learn a metric tensor that faithfully captures the underlying data geometry and informs more accurate flows. We demonstrate the efficacy of this strategy on synthetic manifolds with analytic geodesics, and interpolation of cell |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_25230 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Energy Guided Geometric Flow Matching Zweig, Aaron Zhang, Mingxuan Azizi, Elham Knowles, David Machine Learning Artificial Intelligence A useful inductive bias for temporal data is that trajectories should stay close to the data manifold. Traditional flow matching relies on straight conditional paths, and flow matching methods which learn geodesics rely on RBF kernels or nearest neighbor graphs that suffer from the curse of dimensionality. We propose to use score matching and annealed energy distillation to learn a metric tensor that faithfully captures the underlying data geometry and informs more accurate flows. We demonstrate the efficacy of this strategy on synthetic manifolds with analytic geodesics, and interpolation of cell |
| title | Energy Guided Geometric Flow Matching |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2509.25230 |