• Corpus ID: 120523779

MAX-CSP, Graph Cuts and Statistical Physics

  title={MAX-CSP, Graph Cuts and Statistical Physics},
  author={Yi-Kai Wang},
  journal={arXiv: Data Structures and Algorithms},
  • Yi-Kai Wang
  • Published 30 November 2014
  • Computer Science, Mathematics
  • arXiv: Data Structures and Algorithms
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… 

Figures and Topics from this paper

The Graph Crossing Number and its Variants: A Survey
The crossing number is a popular tool in graph drawing and visualization, but there is not really just one crossing number; there is a large family of crossing number notions of which the crossing


Diameter and Treewidth in Minor-Closed Graph Families
  • D. Eppstein
  • Mathematics, Computer Science
  • 2000
It is shown that treewidth is bounded by a function of the diameter in a minor-closed family, if and only if some apex graph does not belong to the family, and the O(D) bound above can be extended to bounded-genus graphs.
A Simple Algorithm for the Graph Minor Decomposition - Logic meets Structural Graph Theory
A simple quadratic time algorithm for computing graph minor decompositions and combines techniques from logic and structural graph theory and Courcelle's Theorem stating that monadic second-order logic formulas can be evaluated in linear time on graphs of bounded tree width and Robertson and Seymour's so called Weak Structure Theorem.
On the Unique Games Conjecture (Invited Survey)
  • Subhash Khot
  • Mathematics, Computer Science
    2010 IEEE 25th Annual Conference on Computational Complexity
  • 2010
This article surveys recently discovered connections between the Unique Games Conjecture and computational complexity, algorithms, discrete Fourier analysis, and geometry.
Algorithms for Graphs Embeddable with Few Crossings per Edge
It is proved that many optimization problems, including maximum independent set, minimum vertex cover, minimum dominating set and many others, admit polynomial time approximation schemes when restricted to graphs embedded on a surface of bounded genus such that each edge has a bounded number of crossings.
Optimal Inapproximability Results for MAX-CUT and Other 2-Variable CSPs?
This paper shows a reduction from the Unique Games problem to the problem of approximating MAX-CUT to within a factor of $\alpha_{\text{\tiny{GW}}} + \epsilon$ for all $\ep silon > 0$, and indicates that the geometric nature of the Goemans-Williamson algorithm might be intrinsic to the MAX- CUT problem.
Approximation Algorithms for Unit Disk Graphs
A new asymptotic fully-polynomial approximation scheme for unit disk graphs of bounded density is presented, and the scheme for Minimum Connected Dominating Set is the first Baker-like asymPTotic FPTAS for this problem.
"GrabCut": interactive foreground extraction using iterated graph cuts
A more powerful, iterative version of the optimisation of the graph-cut approach is developed and the power of the iterative algorithm is used to simplify substantially the user interaction needed for a given quality of result.
Approximation algorithms for classes of graphs excluding single-crossing graphs as minors
This paper demonstrates structural properties of larger classes of graphs and shows how to exploit the properties to derive polynomial-time algorithms for approximating treewidth and branchwidth and fixed-parameter algorithms for problems such as vertex cover, edge-dominating set, feedback vertex set, and others.
Graph Minors. XX. Wagner's conjecture
We prove Wagner's conjecture, that for every infinite set of finite graphs, one of its members is isomorphic to a minor of another.
Local Tree-Width, Excluded Minors, and Approximation Algorithms
A decomposition theorem for graphs with excluded minors says that such graphs can be decomposed into trees of graphs of almost bounded local tree-width, and it is shown that a number of combinatorial optimization problems have a polynomial time approximation scheme when restricted to a class of graphs with an excluded minor.