Discriminative Gaussian Process Latent Variable Model for Classification
Supervised learning is difficult with high dimensional input spacesand very small training sets, but accurate classification may bepossible if the data lie on a low-dimensional manifold. GaussianProcess Latent Variable Models can discover low dimensional manifoldsgiven only a small number of exampl...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Published: |
2007
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/36901 |
| _version_ | 1826215348834140160 |
|---|---|
| author | Urtasun, Raquel Darrell, Trevor |
| author2 | Trevor Darrell |
| author_facet | Trevor Darrell Urtasun, Raquel Darrell, Trevor |
| author_sort | Urtasun, Raquel |
| collection | MIT |
| description | Supervised learning is difficult with high dimensional input spacesand very small training sets, but accurate classification may bepossible if the data lie on a low-dimensional manifold. GaussianProcess Latent Variable Models can discover low dimensional manifoldsgiven only a small number of examples, but learn a latent spacewithout regard for class labels. Existing methods for discriminativemanifold learning (e.g., LDA, GDA) do constrain the class distributionin the latent space, but are generally deterministic and may notgeneralize well with limited training data. We introduce a method forGaussian Process Classification using latent variable models trainedwith discriminative priors over the latent space, which can learn adiscriminative latent space from a small training set. |
| first_indexed | 2024-09-23T16:25:34Z |
| id | mit-1721.1/36901 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T16:25:34Z |
| publishDate | 2007 |
| record_format | dspace |
| spelling | mit-1721.1/369012019-04-10T09:59:05Z Discriminative Gaussian Process Latent Variable Model for Classification Urtasun, Raquel Darrell, Trevor Trevor Darrell Vision Gaussian Processes Classification Latent Variable Models Machine Learning Supervised learning is difficult with high dimensional input spacesand very small training sets, but accurate classification may bepossible if the data lie on a low-dimensional manifold. GaussianProcess Latent Variable Models can discover low dimensional manifoldsgiven only a small number of examples, but learn a latent spacewithout regard for class labels. Existing methods for discriminativemanifold learning (e.g., LDA, GDA) do constrain the class distributionin the latent space, but are generally deterministic and may notgeneralize well with limited training data. We introduce a method forGaussian Process Classification using latent variable models trainedwith discriminative priors over the latent space, which can learn adiscriminative latent space from a small training set. 2007-03-29T11:21:46Z 2007-03-29T11:21:46Z 2007-03-28 MIT-CSAIL-TR-2007-021 http://hdl.handle.net/1721.1/36901 Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory 8 p. application/postscript application/pdf |
| spellingShingle | Gaussian Processes Classification Latent Variable Models Machine Learning Urtasun, Raquel Darrell, Trevor Discriminative Gaussian Process Latent Variable Model for Classification |
| title | Discriminative Gaussian Process Latent Variable Model for Classification |
| title_full | Discriminative Gaussian Process Latent Variable Model for Classification |
| title_fullStr | Discriminative Gaussian Process Latent Variable Model for Classification |
| title_full_unstemmed | Discriminative Gaussian Process Latent Variable Model for Classification |
| title_short | Discriminative Gaussian Process Latent Variable Model for Classification |
| title_sort | discriminative gaussian process latent variable model for classification |
| topic | Gaussian Processes Classification Latent Variable Models Machine Learning |
| url | http://hdl.handle.net/1721.1/36901 |
| work_keys_str_mv | AT urtasunraquel discriminativegaussianprocesslatentvariablemodelforclassification AT darrelltrevor discriminativegaussianprocesslatentvariablemodelforclassification |