Maximum Likelihood Learning of Conditional MTE Distributions

13 years 11 months ago

Download www.idi.ntnu.no

We describe a procedure for inducing conditional densities within the mixtures of truncated exponentials (MTE) framework. We analyse possible conditional MTE speciﬁcations and propose a model selection scheme, based on the BIC score, for partitioning the domain of the conditioning variables. Finally, experimental results demonstrate the applicability of the learning procedure as well as the expressive power of the conditional MTE distribution.

Helge Langseth, Thomas D. Nielsen, Rafael Rum&iacu

Real-time Traffic

Artificial Intelligence | Conditional Mte | Conditional Mte Distribution | Conditional Mte Speciﬁcations | ECSQARU 2009 |

claim paper

» Parameter estimation and model selection for mixtures of truncated exponentials

» Learning Bayesian network classifiers by maximizing conditional likelihood

» Training Conditional Random Fields Using Virtual Evidence Boosting

» Approximate Parameter Learning in Conditional Random Fields An Empirical Investigation

» Sample Complexity of Composite Likelihood

» A MaximumLikelihood Approach for Multiresolution WOperator Design

» Putting MAP Back on the Map

» Bayesian speaker recognition using Gaussian mixture model and laplace approximation

Post Info
More Details (n/a)

Added	26 May 2010
Updated	26 May 2010
Type	Conference
Year	2009
Where	ECSQARU
Authors	Helge Langseth, Thomas D. Nielsen, Rafael Rumí, Antonio Salmerón

Comments (0)

Sciweavers

Maximum Likelihood Learning of Conditional MTE Distributions

Artificial Intelligence | Conditional Mte | Conditional Mte Distribution | Conditional Mte Speciﬁcations | ECSQARU 2009 |

Explore & Download

Productivity Tools

Document Tools

Image Tools

Sciweavers