Team for Research in
Ubiquitous Secure Technology

Citation
Raquel Urtasun, trevor darrell. "Discriminative Gaussian Process Latent Variable Models for Classification". ICML 2007, June, 2007.

Abstract
Supervised learning is difficult with high dimensional input spaces and very small training sets, but accurate classification may be possible if the data lie on a low-dimensional manifold. Gaussian Process Latent Variable Models can discover low dimensional manifolds given only a small number of examples, but learn a latent space without regard for class labels. Existing methods for discriminative manifold learning (e.g., LDA, GDA) do constrain the class distribution in the latent space, but are generally deterministic and may not generalize well with limited training data. We introduce a method for Gaussian Process Classification using latent variable models trained with discriminative priors over the latent space, which can learn a discriminative latent space from a small training set.

Electronic downloads

• http://www.machinelearning.org/proceedings/icml2007/papers/517.pdf

• HTML

  Raquel Urtasun, trevor darrell. <a
  href="http://www.truststc.org/pubs/277.html"
  >Discriminative Gaussian Process Latent Variable Models
  for Classification</a>, ICML 2007, June, 2007.

• Plain text

  Raquel Urtasun, trevor darrell. "Discriminative
  Gaussian Process Latent Variable Models for
  Classification". ICML 2007, June, 2007.

• BibTeX

  @inproceedings{Urtasundarrell07_DiscriminativeGaussianProcessLatentVariableModelsFor,
      author = {Raquel Urtasun and trevor darrell},
      title = {Discriminative Gaussian Process Latent Variable
                Models for Classification},
      booktitle = {ICML 2007},
      month = {June},
      year = {2007},
      abstract = {Supervised learning is difficult with high
                dimensional input spaces and very small training
                sets, but accurate classification may be possible
                if the data lie on a low-dimensional manifold.
                Gaussian Process Latent Variable Models can
                discover low dimensional manifolds given only a
                small number of examples, but learn a latent space
                without regard for class labels. Existing methods
                for discriminative manifold learning (e.g., LDA,
                GDA) do constrain the class distribution in the
                latent space, but are generally deterministic and
                may not generalize well with limited training
                data. We introduce a method for Gaussian Process
                Classification using latent variable models
                trained with discriminative priors over the latent
                space, which can learn a discriminative latent
                space from a small training set.},
      URL = {http://www.truststc.org/pubs/277.html}
  }
  

Posted by trevor darrell on 30 Jul 2007.
For additional information, see the Publications FAQ or contact webmaster at www truststc org.