Neural Conditional Transport Maps
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908928936247296 |
|---|---|
| author | Rodriguez-Pardo, Carlos Chiani, Leonardo Borgonovo, Emanuele Tavoni, Massimo |
| author_facet | Rodriguez-Pardo, Carlos Chiani, Leonardo Borgonovo, Emanuele Tavoni, Massimo |
| contents | We present a neural framework for learning conditional optimal transport (OT) maps between probability distributions. Our approach introduces a conditioning mechanism capable of processing both categorical and continuous conditioning variables simultaneously. At the core of our method lies a hypernetwork that generates transport layer parameters based on these inputs, creating adaptive mappings that outperform simpler conditioning methods. Comprehensive ablation studies demonstrate the superior performance of our method over baseline configurations. Furthermore, we showcase an application to global sensitivity analysis, offering high performance in computing OT-based sensitivity indices. This work advances the state-of-the-art in conditional optimal transport, enabling broader application of optimal transport principles to complex, high-dimensional domains such as generative modeling and black-box model explainability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_15808 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Neural Conditional Transport Maps Rodriguez-Pardo, Carlos Chiani, Leonardo Borgonovo, Emanuele Tavoni, Massimo Machine Learning Artificial Intelligence Probability Applications 49Q22 (Primary) 68T07 (Secondary) I.5.1; I.2.0; G.3 We present a neural framework for learning conditional optimal transport (OT) maps between probability distributions. Our approach introduces a conditioning mechanism capable of processing both categorical and continuous conditioning variables simultaneously. At the core of our method lies a hypernetwork that generates transport layer parameters based on these inputs, creating adaptive mappings that outperform simpler conditioning methods. Comprehensive ablation studies demonstrate the superior performance of our method over baseline configurations. Furthermore, we showcase an application to global sensitivity analysis, offering high performance in computing OT-based sensitivity indices. This work advances the state-of-the-art in conditional optimal transport, enabling broader application of optimal transport principles to complex, high-dimensional domains such as generative modeling and black-box model explainability. |
| title | Neural Conditional Transport Maps |
| topic | Machine Learning Artificial Intelligence Probability Applications 49Q22 (Primary) 68T07 (Secondary) I.5.1; I.2.0; G.3 |
| url | https://arxiv.org/abs/2505.15808 |