TY - GEN

T1 - Convergent propagation algorithms via oriented trees

AU - Globerson, Amir

AU - Jaakkola, Tommi

N1 - Funding Information:
The work is partially supported by the National Natural Science Foundation of China (Grant NO.61765004, 61465004,61805050,61705050,61535004,61735009, Guangxi Natural Science Foundation 2016GXNSFAA380006, 2017GXNSFAA198164); Guangxi key laboratory of automatic testing technology and instrument (YQ18110).

PY - 2007

Y1 - 2007

N2 - Inference problems in graphical models are often approximated by casting them as constrained optimization problems. Message passing algorithms, such as belief propagation, have previously been suggested as methods for solving these optimization problems. However, there are few convergence guarantees for such algorithms, and the algorithms are therefore not guaranteed to solve the corresponding optimization problem. Here we present an oriented tree decomposition algorithm that is guaranteed to converge to the global optimum of the Tree-Reweighted (TRW) variational problem. Our algorithm performs local updates in the convex dual of the TRW problem - an unconstrained generalized geometric program. Primal updates, also local, correspond to oriented reparametrization operations that leave the distribution intact.

AB - Inference problems in graphical models are often approximated by casting them as constrained optimization problems. Message passing algorithms, such as belief propagation, have previously been suggested as methods for solving these optimization problems. However, there are few convergence guarantees for such algorithms, and the algorithms are therefore not guaranteed to solve the corresponding optimization problem. Here we present an oriented tree decomposition algorithm that is guaranteed to converge to the global optimum of the Tree-Reweighted (TRW) variational problem. Our algorithm performs local updates in the convex dual of the TRW problem - an unconstrained generalized geometric program. Primal updates, also local, correspond to oriented reparametrization operations that leave the distribution intact.

UR - http://www.scopus.com/inward/record.url?scp=80053202574&partnerID=8YFLogxK

M3 - פרסום בספר כנס

AN - SCOPUS:80053202574

SN - 0974903930

SN - 9780974903934

T3 - Proceedings of the 23rd Conference on Uncertainty in Artificial Intelligence, UAI 2007

SP - 133

EP - 140

BT - Proceedings of the 23rd Conference on Uncertainty in Artificial Intelligence, UAI 2007

Y2 - 19 July 2007 through 22 July 2007

ER -