English

Approximate MAP Estimation for Pairwise Potentials via Baker's Technique

Data Structures and Algorithms 2018-04-18 v4

Abstract

The theoretical models providing mathematical abstractions for several significant optimization problems in machine learning, combinatorial optimization, computer vision and statistical physics have intrinsic similarities. We propose a unified framework to model these computation tasks where the structures of these optimization problems are encoded by functions attached on the vertices and edges of a graph. We show that computing MAX 2-CSP admits polynomial-time approximation scheme (PTAS) on planar graphs, graphs with bounded local treewidth, HH-minor-free graphs, geometric graphs with bounded density and graphs embeddable with bounded number of crossings per edge. This implies computing MAX-CUT, MAX-DICUT and MAX kk-CUT admits PTASs on all these classes of graphs. Our method also gives the first PTAS for computing the ground state of ferromagnetic Edwards-Anderson model without external magnetic field on dd-dimensional lattice graphs. These results are widely applicable in vision, graphics and machine learning.

Keywords

Cite

@article{arxiv.1412.0340,
  title  = {Approximate MAP Estimation for Pairwise Potentials via Baker's Technique},
  author = {Yi-Kai Wang},
  journal= {arXiv preprint arXiv:1412.0340},
  year   = {2018}
}
R2 v1 2026-06-22T07:16:25.962Z