STRUCTURE OF A MODEL WITH HIDDEN VARIABLES, METHOD FOR PRODUCING AN INFERENCE FOR SAID MODEL AND METHOD FOR LEARNING THE PARAMETERS FROM DATA FROM SAID MODEL

The invention relates to a data generating structure for a model with hidden variables corresponding to a causal network, wherein the arrows of hidden source variables point to the observed variables. According to the invention, said structure directly allocates continuously observed variables to bi...

Full description

Saved in:
Bibliographic Details
Main Author HAFT, MICHAEL
Format Patent
LanguageEnglish
French
German
Published 26.02.2004
Edition7
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:The invention relates to a data generating structure for a model with hidden variables corresponding to a causal network, wherein the arrows of hidden source variables point to the observed variables. According to the invention, said structure directly allocates continuously observed variables to binary hidden sources.
Bibliography:Application Number: WO2003DE00817