English

A Polyhedral Study of Lifted Multicuts

Discrete Mathematics 2022-02-17 v1 Machine Learning Optimization and Control

Abstract

Fundamental to many applications in data analysis are the decompositions of a graph, i.e. partitions of the node set into component-inducing subsets. One way of encoding decompositions is by multicuts, the subsets of those edges that straddle distinct components. Recently, a lifting of multicuts from a graph G=(V,E)G = (V, E) to an augmented graph G^=(V,EF)\hat G = (V, E \cup F) has been proposed in the field of image analysis, with the goal of obtaining a more expressive characterization of graph decompositions in which it is made explicit also for pairs F(V2)EF \subseteq \tbinom{V}{2} \setminus E of non-neighboring nodes whether these are in the same or distinct components. In this work, we study in detail the polytope in REF\mathbb{R}^{E \cup F} whose vertices are precisely the characteristic vectors of multicuts of G^\hat G lifted from GG, connecting it, in particular, to the rich body of prior work on the clique partitioning and multilinear polytope.

Keywords

Cite

@article{arxiv.2202.08068,
  title  = {A Polyhedral Study of Lifted Multicuts},
  author = {Bjoern Andres and Silvia Di Gregorio and Jannik Irmai and Jan-Hendrik Lange},
  journal= {arXiv preprint arXiv:2202.08068},
  year   = {2022}
}

Comments

63 pages, 18 figures

R2 v1 2026-06-24T09:40:57.066Z