Minimal Inequalities for Constrained Infinite Relaxations of MIPs Sercan Yıldız Carnegie Mellon University, Pittsburgh, PA Introduction

Results

We study the following constrained infinite relaxation of a mixed-integer program: X X x =f + rsr + ryr ,

• Results on Minimal Valid Functions We say that a function π : Rn → R satisfies the generalized symmetry condition if the equality   1  π(r ) = sup  (1 − π(x − f − kr )) : x ∈ S, k ∈ Z++ k holds for all r ∈ Rn. n Theorem: Let π : R → R. π is a minimal valid function for Gf ,S if and only if π(0) = 0, π is subadditive and satisfies the generalized symmetry condition. n n Remark: When S = Z , a subadditive function π : R → R such that π(0) = 0 and 0 ≤ π ≤ 1 satisfies the generalized symmetry condition if n and only if it is periodic with respect to Z and satisfies the symmetry condition. n Theorem: Let ψ, π : R → R be two functions such that (ψ, π) is a valid function pair for Mf ,S . (ψ, π) is minimal for Mf ,S if and only if π is a minimal valid function for Gf ,S and ψ is defined as in (*).

r ∈Rn

(IR)

r ∈Rn n

x ∈ S := P ∩ Z , n sr ∈ R+ for all r ∈ R , n yr ∈ Z+ for all r ∈ R , s, y have finite support. Assumptions: The system (IR) is equivalent to Gomory and Johnson’s mixed-integer infinite group relaxation P is a rational polyhedron. n when S = Z . f ∈ conv(S) \ S. (This is needed only in results on integer lifting.) Let Mf ,S denote the set of solutions (s, y ) that satisfy (IR) for some x ∈ S. n Valid functions. Given two functions ψ, π : R → R, (ψ, π) is said to be valid for Mf ,S if the P P inequality r ∈Rn ψ(r )sr + r ∈Rn π(r )yr ≥ 1 holds for all (s, y ) ∈ Mf ,S . Minimal valid functions. A valid function pair (ψ, π) for Mf ,S is minimal if there is no valid function 0 0 0 0 pair (ψ , π ) distinct from (ψ, π) such that ψ ≤ ψ and π ≤ π. Furthermore, let Gf ,S := {y : (0, y ) ∈ Mf ,S } and Rf ,S := {s : (s, 0) ∈ Mf ,S } represent the set of solutions to the pure-integer and continuous infinite relaxations, respectively.



Results on Integer Lifting

Let Lψ be the lineality space of Rψ . Preliminaries • Preliminaries on Minimal Valid Functions A function π : Rn → R is subadditive if π(r1) + π(r2) ≥ π(r1 + r2) for all r1, r2 ∈ Rn. Given a lattice Λ ⊂ Rn, a function π is periodic with respect to Λ if π(r ) = π(r + w) for all r ∈ Rn and w ∈ Λ. For a n given f ∈ R , a function π is said to satisfy the symmetry condition if π(r ) = 1 − π(−f − r ) for any n r ∈R . n Theorem [GJ72]: Let π : R → R be a nonnegative function. π is a minimal valid function for Gf ,Zn n if and only if π(0) = 0, π is subadditive, periodic with respect to Z and satisfies the symmetry condition. n Theorem [Joh74]: Let ψ, π : R → R be two functions such that (ψ, π) is a valid function pair for Mf ,Zn and π ≥ 0. (ψ, π) is minimal if and only if π is a minimal valid function for Gf ,Zn and ψ is defined as π(r ) n ψ(r ) := lim+ , ∀r ∈ R . (*) →0  • Preliminaries on Integer Lifting n n Given a valid function ψ : R → R for Rf ,S , a function π : R → R is a lifting of ψ if (ψ, π) is valid for Mf ,S . Let ψ be a minimal valid function for Rf ,S . n + Theorem [CCZ11, BCC 12]: There exists a region Rψ ⊂ R , containing the origin in its interior, where any minimal lifting π coincides with ψ. Furthermore, this region is the union of finitely many polyhedra which have the same linear subspace as their recession cone.

lin(conv(S)) = R × {0}. Condition (1) is satisfied in both figures. Condition (2) is satisfied in the right one only.

Theorem: If there exists a unique minimal lifting of ψ, then n Lψ + lin(conv(S)) = R . Corollary: If there exists a unique minimal lifting of ψ, then lin(conv(S)) 6= {0}. Corollary: Suppose Lψ = {0}. If there exists a unique minimal lifting of ψ, n n n then S = Z and Rψ + Z = R . References ˆ M. Conforti, G. Cornuejols, ´ A. Basu, M. Campelo, and G. Zambelli. On lifting integer variables in minimal inequalities. In Proceedings of IPCO 2010, volume 6080 of Lecture Notes in Computer Science, pages 85–95, Lausanne, Switzerland, June 2010. ˆ M. Conforti, G. Cornuejols, ´ A. Basu, M. Campelo, and G. Zambelli. Unique lifting of integer variables. Mathematical Programming Ser. A, 2012. DOI: 10.1007/s10107-012-0560-9. ´ M. Conforti, G. Cornuejols, and G. Zambelli. A geometric perspective on lifting. Operations Research, 59:569–577, 2011.

Theorem [CCZ11]: If Rψ + (lin(conv(S)) ∩ Zn) = Rn, then there exists a unique minimal lifting of ψ.

R.E. Gomory and E.L. Johnson. Some continuous functions related to corner polyhedra. Mathematical Programming, 3:23–85, 1972.

+

Conjecture [BCC 10]: The converse is also true. 2

Rψ (left) and Rψ + Z (right).

´ ´ Images courtesy of Gerard Cornuejols.

Lemma: Rψ + (lin(conv(S)) ∩ Zn) = Rn if and only if n 1 Lψ + lin(conv(S)) = R , and 2 [lin(conv(S)) ∩ Rψ ] + n [lin(conv(S)) ∩ Z ] = lin(conv(S)).

E.L. Johnson. On the group problem for mixed integer programming. Mathematical Programming Study, 2:137–179, 1974. Email: [email protected]

Minimal Inequalities for Constrained Infinite ...

Introduction. We study the following constrained infinite relaxation of a mixed-integer program: x =f + ∑ ... pair (ψ ,π ) distinct from (ψ,π) such that ψ ≤ ψ and π ≤ π. ... function pair for Mf,S . (ψ,π) is minimal for Mf,S if and only if π is a minimal valid function for Gf,S and ψ is defined as in (*). • Results on Integer Lifting. Let L ψ.

178KB Sizes 0 Downloads 264 Views

Recommend Documents

Minimal Inequalities for Constrained Infinite Relaxations of MIPs
We study the following constrained infinite relaxation of a mixed-integer program: x =f + ∑ r∈Rn rsr + ∑ r∈Rn ryr,. (IR) x ∈ S := P ∩ Z n. , sr ∈ R+ for all r ∈ R.

Relative isoperimetric inequalities for minimal ... - Wiley Online Library
Mar 6, 2012 - 10, 1264 – 1273 (2012) / DOI 10.1002/mana.201100078. Relative isoperimetric inequalities for minimal submanifolds outside a convex set. Keomkyo Seo. ∗. Department of Mathematics, Sookmyung Women's University, Hyochangwongil 52, Yong

Minimal Observability for Transactional Hierarchical Services - LaBRI
different level of abstraction, e.g., the top level may describe the interactions ... called a state and t ∈ T a transition) and s0 ∈ Q and sf ∈. Q are the initial and final ..... cal graphs, since a small hierarchical graph can represent an ex

EIGENVALUE ESTIMATES FOR STABLE MINIMAL ...
KEOMKYO SEO. (Communicated by H. Martini). Abstract. In this article ... negative constants, we can extend Theorem 1.1 as follows. THEOREM 1.3. Let M be an ...

OPTIMAL PROBABILITY INEQUALITIES FOR ...
§Department of Discrete Mathematics, Adam Mickiewicz University, Umultowska 87, 61-614. Poznan, Poland ([email protected]). ¶This research was ...

OPTIMAL PROBABILITY INEQUALITIES FOR ...
The inequality is exact and the optimal values of c and k are given explicitly. It improves Kwapien's inequality in the case of the. Rademacher series. We also provide a new and very short proof of the Littlewood-Offord problem without using Sperner'

Volume inequalities for isotropic measures
For each measure Z on Sn−1 the convex body Z∞ ⊆ B may be defined as the convex body whose support function is given by. hZ∞ (u) = max{ u · v : v ∈ supp ...

LIEB–THIRRING INEQUALITIES FOR COMPLEX ...
by Hansmann and Katriel [18] using the complex analytic approach developed in [1]. Their non-selfadjoint version of the Lieb–Thirring inequalities takes the.

Isoperimetric inequalities for submanifolds with ... - Springer Link
Jul 23, 2011 - if ωn is the volume of a unit ball in Rn, then. nnωnVol(D)n−1 ≤ Vol(∂D)n and equality holds if and only if D is a ball. As an extension of the above classical isoperimetric inequality, it is conjectured that any n-dimensional c

Groupwise Constrained Reconstruction for Subspace Clustering
50. 100. 150. 200. 250. Number of Subspaces (Persons). l.h.s.. r.h.s. difference .... an illustration). ..... taining 2 subspaces, each of which contains 50 samples.

Groupwise Constrained Reconstruction for Subspace Clustering - ICML
k=1 dim(Sk). (1). Unfortunately, this assumption will be violated if there exist bases shared among the subspaces. For example, given three orthogonal bases, b1 ...

Groupwise Constrained Reconstruction for Subspace Clustering
The objective of the reconstruction based subspace clustering is to .... Kanade (1998); Kanatani (2001) approximate the data matrix with the ... Analysis (GPCA) (Vidal et al., 2005) fits the samples .... wji and wij could be either small or big.

Google for Education facilitates learning at minimal operating costs for ...
“But computers weren't a natural part of the teaching and they were hardly used in all subjects. Maybe 10% of teachers were using them – and when they did bring them into ... a laptop or an Ipad brought by the student from home. Moreover ...

Groupwise Constrained Reconstruction for Subspace Clustering - ICML
dal, 2009; Liu et al., 2010; Wang et al., 2011). In this paper, we focus .... 2010), Robust Algebraic Segmentation (RAS) is pro- posed to handle the .... fi = det(Ci)− 1. 2 (xi C−1 i xi + νλ). − D+ν. 2. Ci = Hzi − αHxixi. Hk = ∑ j|zj =k

INEQUALITIES BFP.pdf
There was a problem previewing this document. Retrying... Download. Connect more apps... Try one of the apps below to open or edit this item. INEQUALITIES BFP.pdf. INEQUALITIES BFP.pdf. Open. Extract. Open with. Sign In. Main menu. Whoops! There was

Minimal deep flac
image oftheindustry and the FTO membership in the UK and to overseas destination countries 7. Provides ... protection on behalf ofits membership in relation to their nonATOL business 9. ... They can avoid the disadvantage by having broadband where yo

Steiner Minimal Trees - Semantic Scholar
Algorithm: MST-Steiner. Input: A graph G = (V,E,w) and a terminal set L ⊂ V . Output: A Steiner tree T. 1: Construct the metric closure GL on the terminal set L. 2:.

house minimal session.pdf
There was a problem previewing this document. Retrying... Download. Connect more apps... Try one of the apps below to open or edit this item. house minimal ...

A Minimal Article -
Mar 30, 2018 - The sets {i, j, k} and {e1, e2, e3} represent the same set of unit vectors in mutually orthogonal directions in R3. Be warned! We shall use all three ...

Self-Help and Minimal-Contact Therapies for Anxiety ...
form of brief (10–15 min) phone calls or face-to-face meetings wherein the .... (Clark, Kirkby, Daniels, & Marks, 1998), a laptop and desktop computer (Baer, Min-.

A Framework for Minimal Clustering Modification via ...
1https://sites.google.com/site/ .... We experiment with two real world data sets (social network ... ten found clustering simply puts most instances in one clus-.

Minimal Key Lengths for Symmetric Ciphers to Provide ...
Jan 7, 1996 - systems considered adequate for the past several years both fast and cheap. General purpose ... 3MIT Laboratory for Computer Science, [email protected]. 4Counterpane ... 1 Encryption Plays an Essential Role in Protecting the Privacy ..