Generalized sequential tree-reweighted message passing
This provides an incremental improvement for researchers in graphical models, with applications in computer vision and natural language processing.
The paper tackles approximate MAP-MRF inference in general graphical models by generalizing the TRW-S algorithm to handle nested factor constraints, showing experimental improvements over existing methods like min-sum diffusion and MPLP on computer vision and NLP problems.
This paper addresses the problem of approximate MAP-MRF inference in general graphical models. Following [36], we consider a family of linear programming relaxations of the problem where each relaxation is specified by a set of nested pairs of factors for which the marginalization constraint needs to be enforced. We develop a generalization of the TRW-S algorithm [9] for this problem, where we use a decomposition into junction chains, monotonic w.r.t. some ordering on the nodes. This generalizes the monotonic chains in [9] in a natural way. We also show how to deal with nested factors in an efficient way. Experiments show an improvement over min-sum diffusion, MPLP and subgradient ascent algorithms on a number of computer vision and natural language processing problems.