Manifold Matching using Shortest-Path Distance and Joint Neighborhood Selection
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2014
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916300200083456 |
|---|---|
| author | Shen, Cencheng Vogelstein, Joshua T. Priebe, Carey E. |
| author_facet | Shen, Cencheng Vogelstein, Joshua T. Priebe, Carey E. |
| contents | Matching datasets of multiple modalities has become an important task in data analysis. Existing methods often rely on the embedding and transformation of each single modality without utilizing any correspondence information, which often results in sub-optimal matching performance. In this paper, we propose a nonlinear manifold matching algorithm using shortest-path distance and joint neighborhood selection. Specifically, a joint nearest-neighbor graph is built for all modalities. Then the shortest-path distance within each modality is calculated from the joint neighborhood graph, followed by embedding into and matching in a common low-dimensional Euclidean space. Compared to existing algorithms, our approach exhibits superior performance for matching disparate datasets of multiple modalities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_1412_4098 |
| institution | arXiv |
| publishDate | 2014 |
| record_format | arxiv |
| spellingShingle | Manifold Matching using Shortest-Path Distance and Joint Neighborhood Selection Shen, Cencheng Vogelstein, Joshua T. Priebe, Carey E. Machine Learning Matching datasets of multiple modalities has become an important task in data analysis. Existing methods often rely on the embedding and transformation of each single modality without utilizing any correspondence information, which often results in sub-optimal matching performance. In this paper, we propose a nonlinear manifold matching algorithm using shortest-path distance and joint neighborhood selection. Specifically, a joint nearest-neighbor graph is built for all modalities. Then the shortest-path distance within each modality is calculated from the joint neighborhood graph, followed by embedding into and matching in a common low-dimensional Euclidean space. Compared to existing algorithms, our approach exhibits superior performance for matching disparate datasets of multiple modalities. |
| title | Manifold Matching using Shortest-Path Distance and Joint Neighborhood Selection |
| topic | Machine Learning |
| url | https://arxiv.org/abs/1412.4098 |