A method is provided for determining probabilities of states of a system represented by a model including a plurality of nodes connected by links. Each node represents possible states of a corresponding part of the system and each link represents statistical dependencies between possible states of related nodes. The method includes applying a belief propagation algorithm to estimate a minimum energy of the system defining belief propagation messages. The belief propagation messages are compressed and approximate probabilities of the states of the system are determined from the compressed messages.

Title
Efficient message representations for belief propagation algorithms
Application Number
11/962853
Publication Number
20090164192
Application Date
December 21, 2007
Publication Date
June 25, 2009
Inventor
Tianli Yu
Schaumburg
IL, US
Agent
Law Department
IL, US
Motorola
IL, US
Assignee
General Instrument Corporation
PA, US
IPC
G06F 17/50
View Original Source