Unique sink orientations of cubes

  title={Unique sink orientations of cubes},
  author={Tibor Szab{\'o} and Emo Welzl},
  journal={Proceedings 2001 IEEE International Conference on Cluster Computing},
  • Tibor Szabó, E. Welzl
  • Published 14 October 2001
  • Mathematics
  • Proceedings 2001 IEEE International Conference on Cluster Computing
Suppose we are given (the edge graph of) an n-dimensional hypercube with its edges oriented so that every face has a unique sink. [] Key Method We present new algorithms, a deterministic O(1.61/sup n/) procedure and a randomized O((43/20)/sup n/2/)=O(1.47/sup n/) procedure for unique sink orientations. An interesting aspect of these algorithms is that they do not proceed on a path to the sink (in a simplex-like fashion), but they exploit the potential of random access (in the sense of arbitrary access) to…
Unique sink orientations of cubes
The main result is a reduction from linear programming to unique sink orientations, and it turns out that the algorithmic problem is NP-hard only if NP=coNP.
A New Combinatorial Property of Geometric Unique Sink Orientations
A new combinatorial property of the USOs that arise from symmetric P-LCP is established, which includes the US Os that arises from linear and simple convex quadratic programming.
Finding the Sink Takes Some Time: An Almost Quadratic Lower Bound for Finding the Sink of Unique Sink Oriented Cubes
Abstract We give a worst-case Ω(n2/log n) lower bound on the number of vertex evaluations a deterministic algorithm needs to perform in order to find the (unique) sink of a unique sink oriented
Unique sink orientations for homogeneous linear inequalities and their alternative systems
It is not known, at the moment, whether there exists a strongly-polynomial algorithm for solving a system of linear inequalities. A noteworthy candidate is the simplex method, but despite a
Conflict-Free Graph Orientations with Parity Constraints
It is shown that it is NP-complete to decide whether G has an even orientation with exact or subset conflicts, for all conflict sets of size two or higher.
The Complexity of Optimization on Grids
A deterministic algorithm is described showing that the vertex query complexity of d-dimensional grids is O(n^{ \lceil d/2 \rceil }) and for the edge query complexity the authors obtain nearly the same bound, incurring only an $$n^{o(1)}$$no(1) overhead.
Unique Sink Orientations of Grids
It is shown that the orientations arising from PGLCP satisfy the Holt-Klee condition known to hold for polytope digraphs, and the first expected linear-time algorithms for solvingPGLCP with a fixed number of blocks are given.
The Complexity of Recognizing Unique Sink Orientations
This work investigates the connection of recognizing an acyclic USO to PSPACE and proves that the problem is PSPACE-complete, which implies that the straightforward representation of a cycle does not make up for a coNP certificate.
The Niceness of Unique Sink Orientations
Niceness implies natural upper bounds for Random Edge and it is shown that Random Edge is polynomial on at least $n^{\Omega(2^n)}$ many (possibly cyclic) USO.
Worst-case Behaviour of History Based Pivot Rules on Acyclic Unique Sink Orientations of Hypercubes
This paper presents some theoretical and empirical results on the worst case behaviour of various history based pivot rules for this problem, and investigates whether or not they can follow a Hamiltonian path on an acyclic USO.


In a paper presented at a 1996 conference, Holt and Klee introduced a set of necessary conditions for an orientation of the graph of a d-polytope to be induced by a realization into R n and linear
Completely unimodal numberings of a simple polytope
  • K. Hoke
  • Mathematics
    Discret. Appl. Math.
  • 1988
Abstract objective function graphs on the 3-cube
Objective Function Graphs on the cube A Classi cation by Realizability Bernd G artner Volker Kaibel Abstract We call an orientation of the graph of a simple polytope P an abstract objective function
From Linear Separability to Unimodality: A Hierarchy of Pseudo-Boolean Functions
This paper describes a hierarchy of such classes of functions that are shown to strictly contain each other, in increasing order of generality, the threshold, the saddle-free, the pseudomodular, the completely unimodal, the unimmodal, and the unimin functions.
A partition theorem for Euclidean $n$-space
Let Vn be an n-dimensional vector space over the reals. Let s . . ' Sins 7)1, * * I, nbe 2n vectors in Vn such that every sequence of vectors { a,, , an }, where ai is either ti or vi, is a linearly
The Random‐Facet simplex algorithm on combinatorial cubes
  • Bernd Gärtner
  • Computer Science, Mathematics
    Random Struct. Algorithms
  • 2002
The main new result of the paper is a proof that all actual linear programs in Matous̆ek's class are solved by RANDOM‐FACET with an expected polynomial number of $O \bigl(n^2 \bigr)$ pivot steps.
Explicit and Implicit Enforcing - Randomized Optimization
This tutorial is aiming at some randomized approaches for geometric optimization that have been developed over the last ten to fifteen years, and many of the methods can be interpreted as variants of the simplex algorithm, with appropriately chosen randomized pivot rules.
Randomized pivot algorithms for P-matrix linear complementarity problems
  • W. Morris
  • Mathematics, Computer Science
    Math. Program.
  • 2002
It is shown that orientations of the n-cube that come from simple principal pivot algorithms for the linear complementarity problem with a P-matrix properly generalize those that are obtained from linear objective functions on polytopes combinatorially equivalent to the cube.
Combinatorial Linear Programming: Geometry Can Help
This work identifies a "geometric" property of linear programming that goes beyond all abstract notions previously employed in generalized linear programming frameworks, and that can be exploited by the simplex method in a nontrivial setting.
Digraph Models of Bard-Type Algorithms for the Linear Complementarity Problem
These digraphs show that such algorithms based on complementary pivoting for solving the linear complementarity problem can cycle even for symmetric, positive deFinite M, and provide some insight into the algorithms' behavior.