Any2Graph: Deep End-To-End Supervised Graph Prediction With An Optimal Transport Loss
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914972082110464 |
|---|---|
| author | Krzakala, Paul Yang, Junjie Flamary, Rémi d'Alché-Buc, Florence Laclau, Charlotte Labeau, Matthieu |
| author_facet | Krzakala, Paul Yang, Junjie Flamary, Rémi d'Alché-Buc, Florence Laclau, Charlotte Labeau, Matthieu |
| contents | We propose Any2graph, a generic framework for end-to-end Supervised Graph Prediction (SGP) i.e. a deep learning model that predicts an entire graph for any kind of input. The framework is built on a novel Optimal Transport loss, the Partially-Masked Fused Gromov-Wasserstein, that exhibits all necessary properties (permutation invariance, differentiability and scalability) and is designed to handle any-sized graphs. Numerical experiments showcase the versatility of the approach that outperform existing competitors on a novel challenging synthetic dataset and a variety of real-world tasks such as map construction from satellite image (Sat2Graph) or molecule prediction from fingerprint (Fingerprint2Graph). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_12269 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Any2Graph: Deep End-To-End Supervised Graph Prediction With An Optimal Transport Loss Krzakala, Paul Yang, Junjie Flamary, Rémi d'Alché-Buc, Florence Laclau, Charlotte Labeau, Matthieu Machine Learning We propose Any2graph, a generic framework for end-to-end Supervised Graph Prediction (SGP) i.e. a deep learning model that predicts an entire graph for any kind of input. The framework is built on a novel Optimal Transport loss, the Partially-Masked Fused Gromov-Wasserstein, that exhibits all necessary properties (permutation invariance, differentiability and scalability) and is designed to handle any-sized graphs. Numerical experiments showcase the versatility of the approach that outperform existing competitors on a novel challenging synthetic dataset and a variety of real-world tasks such as map construction from satellite image (Sat2Graph) or molecule prediction from fingerprint (Fingerprint2Graph). |
| title | Any2Graph: Deep End-To-End Supervised Graph Prediction With An Optimal Transport Loss |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2402.12269 |