Maximum Likelihood Learning of Conditional MTE Distributions

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.

• Type: Conference
• Year: 2009
• Where: ECSQARU
• Authors: Helge Langseth, Thomas D. Nielsen, Rafael Rumí, Antonio Salmerón
• Keywords: Artificial Intelligence, Conditional Mte, Conditional Mte Distribution, Conditional Mte Specifications, ECSQARU 2009