Join-graph based cost-shifting schemes

10/16/2012
by   Alexander T. Ihler, et al.
0

We develop several algorithms taking advantage of two common approaches for bounding MPE queries in graphical models: minibucket elimination and message-passing updates for linear programming relaxations. Both methods are quite similar, and offer useful perspectives for the other; our hybrid approaches attempt to balance the advantages of each. We demonstrate the power of our hybrid algorithms through extensive empirical evaluation. Most notably, a Branch and Bound search guided by the heuristic function calculated by one of our new algorithms has recently won first place in the PASCAL2 inference challenge.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/04/2012

The Relationship Between AND/OR Search and Variable Elimination

In this paper we compare search and inference in graphical models throug...
research
05/09/2012

MAP Estimation, Message Passing, and Perfect Graphs

Efficiently finding the maximum a posteriori (MAP) configuration of a gr...
research
01/24/2020

Discrete graphical models – an optimization perspective

This monograph is about discrete energy minimization for discrete graphi...
research
11/09/2015

Bayesian Inference in Cumulative Distribution Fields

One approach for constructing copula functions is by multiplication. Giv...
research
03/15/2012

BEEM : Bucket Elimination with External Memory

A major limitation of exact inference algorithms for probabilistic graph...
research
12/12/2012

Iterative Join-Graph Propagation

The paper presents an iterative version of join-tree clustering that app...
research
04/05/2018

Hypertree Decompositions Revisited for PGMs

We revisit the classical problem of exact inference on probabilistic gra...

Please sign up or login with your details

Forgot password? Click here to reset