The Difference Between 5 5 Doubly Nonnegative and Completely Positive Matrices

Size: px
Start display at page:

Download "The Difference Between 5 5 Doubly Nonnegative and Completely Positive Matrices"

Transcription

1 The Difference Between 5 5 Doubly Nonnegative and Completely Positive Matrices Sam Burer and Kurt M. Anstreicher University of Iowa Mirjam Dür TU Darmstadt IMA Hot Topics Workshop, November 2008

2

3 Consider the cones of n n Completely Positive (CPP) and Doubly Nonnegative (DNN) matrices: C n := {X S n : X = NN T for some N 0}, D n := {X S n : X 0, X 0}. We are interested in the difference between C n and D n, particularly for n = 5.

4 Consider the cones of n n Completely Positive (CPP) and Doubly Nonnegative (DNN) matrices: C n := {X S n : X = NN T for some N 0}, D n := {X S n : X 0, X 0}. We are interested in the difference between C n and D n, particularly for n = 5. Why Bother?

5 Consider the cones of n n Completely Positive (CPP) and Doubly Nonnegative (DNN) matrices: C n := {X S n : X = NN T for some N 0}, D n := {X S n : X 0, X 0}. We are interested in the difference between C n and D n, particularly for n = 5. Why Bother? [Burer 2007] Indefinite quadratic programming, with or without binary variables, can be exactly formulated as a linear optimization problem over C n.

6 Consider the cones of n n Completely Positive (CPP) and Doubly Nonnegative (DNN) matrices: C n := {X S n : X = NN T for some N 0}, D n := {X S n : X 0, X 0}. We are interested in the difference between C n and D n, particularly for n = 5. Why Bother? [Burer 2007] Indefinite quadratic programming, with or without binary variables, can be exactly formulated as a linear optimization problem over C n. Optimization over D n is tractable, and clearly C n D n.

7 Consider the cones of n n Completely Positive (CPP) and Doubly Nonnegative (DNN) matrices: C n := {X S n : X = NN T for some N 0}, D n := {X S n : X 0, X 0}. We are interested in the difference between C n and D n, particularly for n = 5. Why Bother? [Burer 2007] Indefinite quadratic programming, with or without binary variables, can be exactly formulated as a linear optimization problem over C n. Optimization over D n is tractable, and clearly C n D n. Known that C n = D n for n 4, but not for n 5.

8 Extreme Rays of D 5 For a symmetric matrix X let G(X) be the undirected graph on vertices {1,..., n} induced by the nonzero off-diagonal elements of X. A complete characterization of extreme rays of D 5 based on rank(x) and G(X) is known. Theorem 1. [Hamilton-Jester and Li, 1996] Suppose X D 5. Then X Ext(D 5 ) if and only if one of the following holds:

9 Extreme Rays of D 5 For a symmetric matrix X let G(X) be the undirected graph on vertices {1,..., n} induced by the nonzero off-diagonal elements of X. A complete characterization of extreme rays of D 5 based on rank(x) and G(X) is known. Theorem 2. [Hamilton-Jester and Li, 1996] Suppose X D 5. Then X Ext(D 5 ) if and only if one of the following holds: (i) rank(x) = 1, in which case X C 5 ;

10 Extreme Rays of D 5 For a symmetric matrix X let G(X) be the undirected graph on vertices {1,..., n} induced by the nonzero off-diagonal elements of X. A complete characterization of extreme rays of D 5 based on rank(x) and G(X) is known. Theorem 3. [Hamilton-Jester and Li, 1996] Suppose X D 5. Then X Ext(D 5 ) if and only if one of the following holds: (i) rank(x) = 1, in which case X C 5 ; (ii) rank(x) = 3 and G(X) is a 5-cycle, in which case X C 5. P T XP =

11 We call X D n \ C n a bad matrix, and X Ext(D n ), X / C n an extremely bad matrix. Let E 5 be the set of extremely bad 5 5 matrices, and B 5 = Cone(E 5 ). It follows that D 5 = C 5 + B 5.

12 We call X D n \ C n a bad matrix, and X Ext(D n ), X / C n an extremely bad matrix. Let E 5 be the set of extremely bad 5 5 matrices, and B 5 = Cone(E 5 ). It follows that D 5 = C 5 + B 5. Can show that B 5 and C 5 have a nontrivial intersection.

13 We call X D n \ C n a bad matrix, and X Ext(D n ), X / C n an extremely bad matrix. Let E 5 be the set of extremely bad 5 5 matrices, and B 5 = Cone(E 5 ). It follows that D 5 = C 5 + B 5. Can show that B 5 and C 5 have a nontrivial intersection. By Caratheodory s Theorem, know that any X D 5 can be represented using at most 16 elements from Ext(D 5 ), any number of which could be in E 5.

14 We call X D n \ C n a bad matrix, and X Ext(D n ), X / C n an extremely bad matrix. Let E 5 be the set of extremely bad 5 5 matrices, and B 5 = Cone(E 5 ). It follows that D 5 = C 5 + B 5. Can show that B 5 and C 5 have a nontrivial intersection. By Caratheodory s Theorem, know that any X D 5 can be represented using at most 16 elements from Ext(D 5 ), any number of which could be in E 5. We prove that in fact D 5 = C 5 + E 5, that is, any X D 5 \ C 5 differs from a CPP matrix by a single extremely bad matrix.

15 CPP Reduction Definition 1. A matrix X D n is CPP-reducible if there are 0 Y C n and Z D n so that X = Y + Z. If no such Y, Z exist then X is CPP-irreducible.

16 CPP Reduction Definition 1. A matrix X D n is CPP-reducible if there are 0 Y C n and Z D n so that X = Y + Z. If no such Y, Z exist then X is CPP-irreducible. Theorem 4. Let X D n. Then X is CPP-reducible iff there exists a partition (I, J) of {1,..., n} such that I, X II > 0, and there is an f Range(X) with f 0, f J = 0, f I 0.

17 CPP Reduction Definition 1. A matrix X D n is CPP-reducible if there are 0 Y C n and Z D n so that X = Y + Z. If no such Y, Z exist then X is CPP-irreducible. Theorem 4. Let X D n. Then X is CPP-reducible iff there exists a partition (I, J) of {1,..., n} such that I, X II > 0, and there is an f Range(X) with f 0, f J = 0, f I 0. For any I, conditions in Theorem 4 can be checked using the LP max e T f s. t. Xw = f, e T f 1, (1) f 0, f J = 0. Let f be an optimal solution. If e T f = 1 then X is CPPreducible, and can take Y := ε f (f ) T and Z := X ε f (f ) T, where ε = argmax{ε > 0 : X εf (f ) T D n }.

18 Algorithm 1 CPP Reduction of a DNN Matrix Input: X D n 1: Set Y 0 := 0 and Z 0 := X. 2: for k = 0, 1, 2,... do 3: Find a partition (I, J) of {1,..., n} such that: (i) I, (ii) [Z k ] II > 0, (iii) the optimal value of (1) with X replaced by Z k is one. Let f k be an optimal solution of (1). If no such partition, set Y := Y k, Z := Z k and STOP. 4: Let ε k = argmax{ε > 0 : Z k εf k fk T D n}. 5: Set Y k+1 := Y k + ε k f k fk T and Z k+1 := Z k ε k f k fk T. 6: end for 7: Set Y := Y k+1 and Z := Z k+1. Output: Y C n and Z D n with X = Y + Z.

19 Theorem 5. Algorithm 1 terminates after no more than n + n(n + 1)/2 iterations.

20 Theorem 5. Algorithm 1 terminates after no more than n + n(n + 1)/2 iterations. Algorithm 1 is not polynomial-time; number of partitions that must be checked is exponential in n.

21 Theorem 5. Algorithm 1 terminates after no more than n + n(n + 1)/2 iterations. Algorithm 1 is not polynomial-time; number of partitions that must be checked is exponential in n. In decomposition X = Y + Z produced by Algorithm 1, Z D n is CPP-irreducible.

22 Theorem 5. Algorithm 1 terminates after no more than n + n(n + 1)/2 iterations. Algorithm 1 is not polynomial-time; number of partitions that must be checked is exponential in n. In decomposition X = Y + Z produced by Algorithm 1, Z D n is CPP-irreducible. If input to Algorithm 1 is X C n, output could be X = Y + Z where 0 Z D n. In particular Algorithm 1 does not determine if a given input X is CPP.

23 Theorem 5. Algorithm 1 terminates after no more than n + n(n + 1)/2 iterations. Algorithm 1 is not polynomial-time; number of partitions that must be checked is exponential in n. In decomposition X = Y + Z produced by Algorithm 1, Z D n is CPP-irreducible. If input to Algorithm 1 is X C n, output could be X = Y + Z where 0 Z D n. In particular Algorithm 1 does not determine if a given input X is CPP. Theorem 6. Let X D 5. Then X is CPP-irreducible if and only if X is extremely bad. Moreover, if X D 5 \ C 5 then there are Y C 5 and Z E 5 so that X = Y + Z.

24 Proof of Theorem 6 uses new explicit representation for matrices in E 5 : Theorem 7. X E 5 if only if there exists a permutation matrix P, a positive-diagonal matrix Λ and a 5 3 matrix R = r 21 r r r 21 (r 21 > 0, r 22 > 0) such that P T XP = ΛRR T Λ.

25 Separating an X E 5 from C 5 We also use Theorem 7 to construct a cut that separates a given X E 5 from C 5. Let H be the Horn matrix H := H is a copositive matrix which cannot be represented as the sum of a positive semidefinite and a nonnegative matrix, i.e., H C 5 \D 5.

26 Separating an X E 5 from C 5 We also use Theorem 7 to construct a cut that separates a given X E 5 from C 5. Let H be the Horn matrix H := H is a copositive matrix which cannot be represented as the sum of a positive semidefinite and a nonnegative matrix, i.e., H C 5 \D 5. Theorem 8. Let X E 5 with P T XP = ΛRR T Λ its representation provided by Theorem 7. Define w := (P T XP H) 1 e > 0 and K := P Diag(w)H Diag(w)P T. Then K Y 0 for all Y C 5, but K X < 0.

27 Application: Computing the maximum stable set in a graph. Let A be the adacency matrix of a graph G on n vertices, and let α be the maximum size of a stable set. It is known that α 1 = min {(I + A) X : E X = 1, X C n }. (2)

28 Application: Computing the maximum stable set in a graph. Let A be the adacency matrix of a graph G on n vertices, and let α be the maximum size of a stable set. It is known that α 1 = min {(I + A) X : E X = 1, X C n }. (2) Relaxing C n to D n results in a polynomial-time computable upper bound on α: (ϑ ) 1 = min {(I + A) X : E X = 1, X D n }. (3) The bound ϑ was first established (via a different derivation) by Schrijver as a strengthening of Lovász s ϑ number.

29 Application: Computing the maximum stable set in a graph. Let A be the adacency matrix of a graph G on n vertices, and let α be the maximum size of a stable set. It is known that α 1 = min {(I + A) X : E X = 1, X C n }. (2) Relaxing C n to D n results in a polynomial-time computable upper bound on α: (ϑ ) 1 = min {(I + A) X : E X = 1, X D n }. (3) The bound ϑ was first established (via a different derivation) by Schrijver as a strengthening of Lovász s ϑ number. For the case where G is a 5-cycle have α = 2 but ϑ = 5 > 2.

30 Application: Computing the maximum stable set in a graph. Let A be the adacency matrix of a graph G on n vertices, and let α be the maximum size of a stable set. It is known that α 1 = min {(I + A) X : E X = 1, X C n }. (2) Relaxing C n to D n results in a polynomial-time computable upper bound on α: (ϑ ) 1 = min {(I + A) X : E X = 1, X D n }. (3) The bound ϑ was first established (via a different derivation) by Schrijver as a strengthening of Lovász s ϑ number. For the case where G is a 5-cycle have α = 2 but ϑ = 5 > 2. Unique solution of problem in (3) is an X E 5. Adding cut from Theorem 8 and re-solving inc reases solution value from 1/ 5 to 1/2; i.e. gap between (3) and (2) is closed.

31 Open Problem 1: Separate arbitrary X D 5 \ C 5 from C 5.

32 Open Problem 1: Separate arbitrary X D 5 \ C 5 from C 5. Know that X D 5 \ C 5 has a decomposition X = Y + Z, Y C 5, Z E 5.

33 Open Problem 1: Separate arbitrary X D 5 \ C 5 from C 5. Know that X D 5 \ C 5 has a decomposition X = Y + Z, Y C 5, Z E 5. Separation procedure described above covers the case where Y = 0.

34 Open Problem 1: Separate arbitrary X D 5 \ C 5 from C 5. Know that X D 5 \ C 5 has a decomposition X = Y + Z, Y C 5, Z E 5. Separation procedure described above covers the case where Y = 0. If Y 0, separation cannot be based on representation X = Y + Z alone, since know that X C 5 may have representation of this form with Z 0.

35 Open Problem 1: Separate arbitrary X D 5 \ C 5 from C 5. Know that X D 5 \ C 5 has a decomposition X = Y + Z, Y C 5, Z E 5. Separation procedure described above covers the case where Y = 0. If Y 0, separation cannot be based on representation X = Y + Z alone, since know that X C 5 may have representation of this form with Z 0. Open Problem 2: Give complete inner description of C 5 give complete outer description of C 5.

36

Linear Algebra and its Applications

Linear Algebra and its Applications Linear Algebra and its Applications 431 (2009) 1539 1552 Contents lists available at ScienceDirect Linear Algebra and its Applications journal homepage: www.elsevier.com/locate/laa The difference between

More information

Separating Doubly Nonnegative and Completely Positive Matrices

Separating Doubly Nonnegative and Completely Positive Matrices Separating Doubly Nonnegative and Completely Positive Matrices Hongbo Dong and Kurt Anstreicher March 8, 2010 Abstract The cone of Completely Positive (CP) matrices can be used to exactly formulate a variety

More information

On the structure of the 5 x 5 copositive cone

On the structure of the 5 x 5 copositive cone On the structure of the 5 x 5 copositive cone Roland Hildebrand 1 Mirjam Dür 2 Peter Dickinson Luuk Gijbens 3 1 Laboratory Jean Kuntzmann, University Grenoble 1 / CNRS 2 Mathematics Dept., University Trier

More information

Copositive Programming and Combinatorial Optimization

Copositive Programming and Combinatorial Optimization Copositive Programming and Combinatorial Optimization Franz Rendl http://www.math.uni-klu.ac.at Alpen-Adria-Universität Klagenfurt Austria joint work with I.M. Bomze (Wien) and F. Jarre (Düsseldorf) IMA

More information

A note on 5 5 Completely positive matrices

A note on 5 5 Completely positive matrices A note on 5 5 Completely positive matrices Hongbo Dong and Kurt Anstreicher October 2009; revised April 2010 Abstract In their paper 5 5 Completely positive matrices, Berman and Xu [BX04] attempt to characterize

More information

Copositive Programming and Combinatorial Optimization

Copositive Programming and Combinatorial Optimization Copositive Programming and Combinatorial Optimization Franz Rendl http://www.math.uni-klu.ac.at Alpen-Adria-Universität Klagenfurt Austria joint work with M. Bomze (Wien) and F. Jarre (Düsseldorf) and

More information

Scaling relationship between the copositive cone and Parrilo s first level approximation

Scaling relationship between the copositive cone and Parrilo s first level approximation Scaling relationship between the copositive cone and Parrilo s first level approximation Peter J.C. Dickinson University of Groningen University of Vienna University of Twente Mirjam Dür University of

More information

The maximal stable set problem : Copositive programming and Semidefinite Relaxations

The maximal stable set problem : Copositive programming and Semidefinite Relaxations The maximal stable set problem : Copositive programming and Semidefinite Relaxations Kartik Krishnan Department of Mathematical Sciences Rensselaer Polytechnic Institute Troy, NY 12180 USA kartis@rpi.edu

More information

Approximation Algorithms

Approximation Algorithms Approximation Algorithms Chapter 26 Semidefinite Programming Zacharias Pitouras 1 Introduction LP place a good lower bound on OPT for NP-hard problems Are there other ways of doing this? Vector programs

More information

CSC Linear Programming and Combinatorial Optimization Lecture 12: The Lift and Project Method

CSC Linear Programming and Combinatorial Optimization Lecture 12: The Lift and Project Method CSC2411 - Linear Programming and Combinatorial Optimization Lecture 12: The Lift and Project Method Notes taken by Stefan Mathe April 28, 2007 Summary: Throughout the course, we have seen the importance

More information

Interior points of the completely positive cone

Interior points of the completely positive cone Electronic Journal of Linear Algebra Volume 17 Volume 17 (2008) Article 5 2008 Interior points of the completely positive cone Mirjam Duer duer@mathematik.tu-darmstadt.de Georg Still Follow this and additional

More information

Nonconvex Quadratic Programming: Return of the Boolean Quadric Polytope

Nonconvex Quadratic Programming: Return of the Boolean Quadric Polytope Nonconvex Quadratic Programming: Return of the Boolean Quadric Polytope Kurt M. Anstreicher Dept. of Management Sciences University of Iowa Seminar, Chinese University of Hong Kong, October 2009 We consider

More information

Rank-one Generated Spectral Cones Defined by Two Homogeneous Linear Matrix Inequalities

Rank-one Generated Spectral Cones Defined by Two Homogeneous Linear Matrix Inequalities Rank-one Generated Spectral Cones Defined by Two Homogeneous Linear Matrix Inequalities C.J. Argue Joint work with Fatma Kılınç-Karzan October 22, 2017 INFORMS Annual Meeting Houston, Texas C. Argue, F.

More information

Comparing Convex Relaxations for Quadratically Constrained Quadratic Programming

Comparing Convex Relaxations for Quadratically Constrained Quadratic Programming Comparing Convex Relaxations for Quadratically Constrained Quadratic Programming Kurt M. Anstreicher Dept. of Management Sciences University of Iowa European Workshop on MINLP, Marseille, April 2010 The

More information

A Geometrical Analysis of a Class of Nonconvex Conic Programs for Convex Conic Reformulations of Quadratic and Polynomial Optimization Problems

A Geometrical Analysis of a Class of Nonconvex Conic Programs for Convex Conic Reformulations of Quadratic and Polynomial Optimization Problems A Geometrical Analysis of a Class of Nonconvex Conic Programs for Convex Conic Reformulations of Quadratic and Polynomial Optimization Problems Sunyoung Kim, Masakazu Kojima, Kim-Chuan Toh arxiv:1901.02179v1

More information

Semidefinite and Second Order Cone Programming Seminar Fall 2001 Lecture 5

Semidefinite and Second Order Cone Programming Seminar Fall 2001 Lecture 5 Semidefinite and Second Order Cone Programming Seminar Fall 2001 Lecture 5 Instructor: Farid Alizadeh Scribe: Anton Riabov 10/08/2001 1 Overview We continue studying the maximum eigenvalue SDP, and generalize

More information

Relations between Semidefinite, Copositive, Semi-infinite and Integer Programming

Relations between Semidefinite, Copositive, Semi-infinite and Integer Programming Relations between Semidefinite, Copositive, Semi-infinite and Integer Programming Author: Faizan Ahmed Supervisor: Dr. Georg Still Master Thesis University of Twente the Netherlands May 2010 Relations

More information

Copositive matrices and periodic dynamical systems

Copositive matrices and periodic dynamical systems Extreme copositive matrices and periodic dynamical systems Weierstrass Institute (WIAS), Berlin Optimization without borders Dedicated to Yuri Nesterovs 60th birthday February 11, 2016 and periodic dynamical

More information

Convex relaxation. In example below, we have N = 6, and the cut we are considering

Convex relaxation. In example below, we have N = 6, and the cut we are considering Convex relaxation The art and science of convex relaxation revolves around taking a non-convex problem that you want to solve, and replacing it with a convex problem which you can actually solve the solution

More information

A solution approach for linear optimization with completely positive matrices

A solution approach for linear optimization with completely positive matrices A solution approach for linear optimization with completely positive matrices Franz Rendl http://www.math.uni-klu.ac.at Alpen-Adria-Universität Klagenfurt Austria joint work with M. Bomze (Wien) and F.

More information

Convex relaxation. In example below, we have N = 6, and the cut we are considering

Convex relaxation. In example below, we have N = 6, and the cut we are considering Convex relaxation The art and science of convex relaxation revolves around taking a non-convex problem that you want to solve, and replacing it with a convex problem which you can actually solve the solution

More information

Lift-and-Project Techniques and SDP Hierarchies

Lift-and-Project Techniques and SDP Hierarchies MFO seminar on Semidefinite Programming May 30, 2010 Typical combinatorial optimization problem: max c T x s.t. Ax b, x {0, 1} n P := {x R n Ax b} P I := conv(k {0, 1} n ) LP relaxation Integral polytope

More information

The extreme rays of the 5 5 copositive cone

The extreme rays of the 5 5 copositive cone The extreme rays of the copositive cone Roland Hildebrand March 8, 0 Abstract We give an explicit characterization of all extreme rays of the cone C of copositive matrices. The results are based on the

More information

An Adaptive Linear Approximation Algorithm for Copositive Programs

An Adaptive Linear Approximation Algorithm for Copositive Programs 1 An Adaptive Linear Approximation Algorithm for Copositive Programs Stefan Bundfuss and Mirjam Dür 1 Department of Mathematics, Technische Universität Darmstadt, Schloßgartenstr. 7, D 64289 Darmstadt,

More information

arxiv: v1 [math.oc] 23 Nov 2012

arxiv: v1 [math.oc] 23 Nov 2012 arxiv:1211.5406v1 [math.oc] 23 Nov 2012 The equivalence between doubly nonnegative relaxation and semidefinite relaxation for binary quadratic programming problems Abstract Chuan-Hao Guo a,, Yan-Qin Bai

More information

An improved characterisation of the interior of the completely positive cone

An improved characterisation of the interior of the completely positive cone Electronic Journal of Linear Algebra Volume 2 Volume 2 (2) Article 5 2 An improved characterisation of the interior of the completely positive cone Peter J.C. Dickinson p.j.c.dickinson@rug.nl Follow this

More information

Applications of semidefinite programming in Algebraic Combinatorics

Applications of semidefinite programming in Algebraic Combinatorics Applications of semidefinite programming in Algebraic Combinatorics Tohoku University The 23rd RAMP Symposium October 24, 2011 We often want to 1 Bound the value of a numerical parameter of certain combinatorial

More information

Modeling with semidefinite and copositive matrices

Modeling with semidefinite and copositive matrices Modeling with semidefinite and copositive matrices Franz Rendl http://www.math.uni-klu.ac.at Alpen-Adria-Universität Klagenfurt Austria F. Rendl, Singapore workshop 2006 p.1/24 Overview Node and Edge relaxations

More information

Considering Copositivity Locally

Considering Copositivity Locally Considering Copositivity Locally Peter J.C. Dickinson Uni. of Groningen Uni. of Vienna Uni. of Twente Roland Hildebrand Uni. of Grenoble Weierstrass Institute Uni. of Grenoble IFORS 2017 Thursday 20th

More information

Hypergraph Matching by Linear and Semidefinite Programming. Yves Brise, ETH Zürich, Based on 2010 paper by Chan and Lau

Hypergraph Matching by Linear and Semidefinite Programming. Yves Brise, ETH Zürich, Based on 2010 paper by Chan and Lau Hypergraph Matching by Linear and Semidefinite Programming Yves Brise, ETH Zürich, 20110329 Based on 2010 paper by Chan and Lau Introduction Vertex set V : V = n Set of hyperedges E Hypergraph matching:

More information

Copositive Plus Matrices

Copositive Plus Matrices Copositive Plus Matrices Willemieke van Vliet Master Thesis in Applied Mathematics October 2011 Copositive Plus Matrices Summary In this report we discuss the set of copositive plus matrices and their

More information

The Trust Region Subproblem with Non-Intersecting Linear Constraints

The Trust Region Subproblem with Non-Intersecting Linear Constraints The Trust Region Subproblem with Non-Intersecting Linear Constraints Samuel Burer Boshi Yang February 21, 2013 Abstract This paper studies an extended trust region subproblem (etrs in which the trust region

More information

Semidefinite Programming Basics and Applications

Semidefinite Programming Basics and Applications Semidefinite Programming Basics and Applications Ray Pörn, principal lecturer Åbo Akademi University Novia University of Applied Sciences Content What is semidefinite programming (SDP)? How to represent

More information

The Steiner Network Problem

The Steiner Network Problem The Steiner Network Problem Pekka Orponen T-79.7001 Postgraduate Course on Theoretical Computer Science 7.4.2008 Outline 1. The Steiner Network Problem Linear programming formulation LP relaxation 2. The

More information

Lagrangian-Conic Relaxations, Part I: A Unified Framework and Its Applications to Quadratic Optimization Problems

Lagrangian-Conic Relaxations, Part I: A Unified Framework and Its Applications to Quadratic Optimization Problems Lagrangian-Conic Relaxations, Part I: A Unified Framework and Its Applications to Quadratic Optimization Problems Naohiko Arima, Sunyoung Kim, Masakazu Kojima, and Kim-Chuan Toh Abstract. In Part I of

More information

CSCI 1951-G Optimization Methods in Finance Part 10: Conic Optimization

CSCI 1951-G Optimization Methods in Finance Part 10: Conic Optimization CSCI 1951-G Optimization Methods in Finance Part 10: Conic Optimization April 6, 2018 1 / 34 This material is covered in the textbook, Chapters 9 and 10. Some of the materials are taken from it. Some of

More information

6-1 The Positivstellensatz P. Parrilo and S. Lall, ECC

6-1 The Positivstellensatz P. Parrilo and S. Lall, ECC 6-1 The Positivstellensatz P. Parrilo and S. Lall, ECC 2003 2003.09.02.10 6. The Positivstellensatz Basic semialgebraic sets Semialgebraic sets Tarski-Seidenberg and quantifier elimination Feasibility

More information

Advanced Mathematical Programming IE417. Lecture 24. Dr. Ted Ralphs

Advanced Mathematical Programming IE417. Lecture 24. Dr. Ted Ralphs Advanced Mathematical Programming IE417 Lecture 24 Dr. Ted Ralphs IE417 Lecture 24 1 Reading for This Lecture Sections 11.2-11.2 IE417 Lecture 24 2 The Linear Complementarity Problem Given M R p p and

More information

Lecture Semidefinite Programming and Graph Partitioning

Lecture Semidefinite Programming and Graph Partitioning Approximation Algorithms and Hardness of Approximation April 16, 013 Lecture 14 Lecturer: Alantha Newman Scribes: Marwa El Halabi 1 Semidefinite Programming and Graph Partitioning In previous lectures,

More information

Downloaded 07/16/12 to Redistribution subject to SIAM license or copyright; see

Downloaded 07/16/12 to Redistribution subject to SIAM license or copyright; see SIAM J. MATRIX ANAL. APPL. Vol. 33, No. 3, pp. 71 72 c 212 Society for Industrial and Applied Mathematics LINEAR-TIME COMPLETE POSITIVITY DETECTION AND DECOMPOSITION OF SPARSE MATRICES PETER J. C. DICKINSON

More information

Research Reports on Mathematical and Computing Sciences

Research Reports on Mathematical and Computing Sciences ISSN 1342-2804 Research Reports on Mathematical and Computing Sciences Doubly Nonnegative Relaxations for Quadratic and Polynomial Optimization Problems with Binary and Box Constraints Sunyoung Kim, Masakazu

More information

Proof Complexity Meets Algebra

Proof Complexity Meets Algebra ICALP 17, Warsaw 11th July 2017 (CSP problem) P 3-COL S resolution (proof system) Proofs in S of the fact that an instance of P is unsatisfiable. Resolution proofs of a graph being not 3-colorable. Standard

More information

University of Groningen. Copositive Programming a Survey Dür, Mirjam. Published in: EPRINTS-BOOK-TITLE

University of Groningen. Copositive Programming a Survey Dür, Mirjam. Published in: EPRINTS-BOOK-TITLE University of Groningen Copositive Programming a Survey Dür, Mirjam Published in: EPRINTS-BOOK-TITLE IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to

More information

Introduction to Semidefinite Programming I: Basic properties a

Introduction to Semidefinite Programming I: Basic properties a Introduction to Semidefinite Programming I: Basic properties and variations on the Goemans-Williamson approximation algorithm for max-cut MFO seminar on Semidefinite Programming May 30, 2010 Semidefinite

More information

SDP Relaxations for MAXCUT

SDP Relaxations for MAXCUT SDP Relaxations for MAXCUT from Random Hyperplanes to Sum-of-Squares Certificates CATS @ UMD March 3, 2017 Ahmed Abdelkader MAXCUT SDP SOS March 3, 2017 1 / 27 Overview 1 MAXCUT, Hardness and UGC 2 LP

More information

Dimension reduction for semidefinite programming

Dimension reduction for semidefinite programming 1 / 22 Dimension reduction for semidefinite programming Pablo A. Parrilo Laboratory for Information and Decision Systems Electrical Engineering and Computer Science Massachusetts Institute of Technology

More information

WHEN DOES THE POSITIVE SEMIDEFINITENESS CONSTRAINT HELP IN LIFTING PROCEDURES?

WHEN DOES THE POSITIVE SEMIDEFINITENESS CONSTRAINT HELP IN LIFTING PROCEDURES? MATHEMATICS OF OPERATIONS RESEARCH Vol. 6, No. 4, November 00, pp. 796 85 Printed in U.S.A. WHEN DOES THE POSITIVE SEMIDEFINITENESS CONSTRAINT HELP IN LIFTING PROCEDURES? MICHEL X. GOEMANS and LEVENT TUNÇEL

More information

Representations of All Solutions of Boolean Programming Problems

Representations of All Solutions of Boolean Programming Problems Representations of All Solutions of Boolean Programming Problems Utz-Uwe Haus and Carla Michini Institute for Operations Research Department of Mathematics ETH Zurich Rämistr. 101, 8092 Zürich, Switzerland

More information

47-831: Advanced Integer Programming Lecturer: Amitabh Basu Lecture 2 Date: 03/18/2010

47-831: Advanced Integer Programming Lecturer: Amitabh Basu Lecture 2 Date: 03/18/2010 47-831: Advanced Integer Programming Lecturer: Amitabh Basu Lecture Date: 03/18/010 We saw in the previous lecture that a lattice Λ can have many bases. In fact, if Λ is a lattice of a subspace L with

More information

Cuts for mixed 0-1 conic programs

Cuts for mixed 0-1 conic programs Cuts for mixed 0-1 conic programs G. Iyengar 1 M. T. Cezik 2 1 IEOR Department Columbia University, New York. 2 GERAD Université de Montréal, Montréal TU-Chemnitz Workshop on Integer Programming and Continuous

More information

A notion of Total Dual Integrality for Convex, Semidefinite and Extended Formulations

A notion of Total Dual Integrality for Convex, Semidefinite and Extended Formulations A notion of for Convex, Semidefinite and Extended Formulations Marcel de Carli Silva Levent Tunçel April 26, 2018 A vector in R n is integral if each of its components is an integer, A vector in R n is

More information

A Note on Representations of Linear Inequalities in Non-Convex Mixed-Integer Quadratic Programs

A Note on Representations of Linear Inequalities in Non-Convex Mixed-Integer Quadratic Programs A Note on Representations of Linear Inequalities in Non-Convex Mixed-Integer Quadratic Programs Adam N. Letchford Daniel J. Grainger To appear in Operations Research Letters Abstract In the literature

More information

Fast ADMM for Sum of Squares Programs Using Partial Orthogonality

Fast ADMM for Sum of Squares Programs Using Partial Orthogonality Fast ADMM for Sum of Squares Programs Using Partial Orthogonality Antonis Papachristodoulou Department of Engineering Science University of Oxford www.eng.ox.ac.uk/control/sysos antonis@eng.ox.ac.uk with

More information

On Valid Inequalities for Quadratic Programming with Continuous Variables and Binary Indicators

On Valid Inequalities for Quadratic Programming with Continuous Variables and Binary Indicators On Valid Inequalities for Quadratic Programming with Continuous Variables and Binary Indicators Hongbo Dong and Jeff Linderoth Wisconsin Institutes for Discovery University of Wisconsin-Madison, USA, hdong6,linderoth}@wisc.edu

More information

New Lower Bounds on the Stability Number of a Graph

New Lower Bounds on the Stability Number of a Graph New Lower Bounds on the Stability Number of a Graph E. Alper Yıldırım June 27, 2007 Abstract Given a simple, undirected graph G, Motzkin and Straus [Canadian Journal of Mathematics, 17 (1965), 533 540]

More information

Analysis of Copositive Optimization Based Linear Programming Bounds on Standard Quadratic Optimization

Analysis of Copositive Optimization Based Linear Programming Bounds on Standard Quadratic Optimization Analysis of Copositive Optimization Based Linear Programming Bounds on Standard Quadratic Optimization Gizem Sağol E. Alper Yıldırım April 18, 2014 Abstract The problem of minimizing a quadratic form over

More information

Sparse Matrix Theory and Semidefinite Optimization

Sparse Matrix Theory and Semidefinite Optimization Sparse Matrix Theory and Semidefinite Optimization Lieven Vandenberghe Department of Electrical Engineering University of California, Los Angeles Joint work with Martin S. Andersen and Yifan Sun Third

More information

Travelling Salesman Problem

Travelling Salesman Problem Travelling Salesman Problem Fabio Furini November 10th, 2014 Travelling Salesman Problem 1 Outline 1 Traveling Salesman Problem Separation Travelling Salesman Problem 2 (Asymmetric) Traveling Salesman

More information

Math Matrix Algebra

Math Matrix Algebra Math 44 - Matrix Algebra Review notes - (Alberto Bressan, Spring 7) sec: Orthogonal diagonalization of symmetric matrices When we seek to diagonalize a general n n matrix A, two difficulties may arise:

More information

Iowa State University and American Institute of Mathematics

Iowa State University and American Institute of Mathematics Matrix and Matrix and Iowa State University and American Institute of Mathematics ICART 2008 May 28, 2008 Inverse Eigenvalue Problem for a Graph () Problem for a Graph of minimum rank Specific matrices

More information

Technische Universität Ilmenau Institut für Mathematik

Technische Universität Ilmenau Institut für Mathematik Technische Universität Ilmenau Institut für Mathematik Preprint No. M 14/05 Copositivity tests based on the linear complementarity problem Carmo Bras, Gabriele Eichfelder and Joaquim Judice 28. August

More information

LINEAR SYSTEMS (11) Intensive Computation

LINEAR SYSTEMS (11) Intensive Computation LINEAR SYSTEMS () Intensive Computation 27-8 prof. Annalisa Massini Viviana Arrigoni EXACT METHODS:. GAUSSIAN ELIMINATION. 2. CHOLESKY DECOMPOSITION. ITERATIVE METHODS:. JACOBI. 2. GAUSS-SEIDEL 2 CHOLESKY

More information

Lecture Note 5: Semidefinite Programming for Stability Analysis

Lecture Note 5: Semidefinite Programming for Stability Analysis ECE7850: Hybrid Systems:Theory and Applications Lecture Note 5: Semidefinite Programming for Stability Analysis Wei Zhang Assistant Professor Department of Electrical and Computer Engineering Ohio State

More information

arxiv: v3 [math.ra] 10 Jun 2016

arxiv: v3 [math.ra] 10 Jun 2016 To appear in Linear and Multilinear Algebra Vol. 00, No. 00, Month 0XX, 1 10 The critical exponent for generalized doubly nonnegative matrices arxiv:1407.7059v3 [math.ra] 10 Jun 016 Xuchen Han a, Charles

More information

Cutting Planes for First Level RLT Relaxations of Mixed 0-1 Programs

Cutting Planes for First Level RLT Relaxations of Mixed 0-1 Programs Cutting Planes for First Level RLT Relaxations of Mixed 0-1 Programs 1 Cambridge, July 2013 1 Joint work with Franklin Djeumou Fomeni and Adam N. Letchford Outline 1. Introduction 2. Literature Review

More information

Nonnegative Matrices I

Nonnegative Matrices I Nonnegative Matrices I Daisuke Oyama Topics in Economic Theory September 26, 2017 References J. L. Stuart, Digraphs and Matrices, in Handbook of Linear Algebra, Chapter 29, 2006. R. A. Brualdi and H. J.

More information

LMI MODELLING 4. CONVEX LMI MODELLING. Didier HENRION. LAAS-CNRS Toulouse, FR Czech Tech Univ Prague, CZ. Universidad de Valladolid, SP March 2009

LMI MODELLING 4. CONVEX LMI MODELLING. Didier HENRION. LAAS-CNRS Toulouse, FR Czech Tech Univ Prague, CZ. Universidad de Valladolid, SP March 2009 LMI MODELLING 4. CONVEX LMI MODELLING Didier HENRION LAAS-CNRS Toulouse, FR Czech Tech Univ Prague, CZ Universidad de Valladolid, SP March 2009 Minors A minor of a matrix F is the determinant of a submatrix

More information

Hilbert s 17th Problem to Semidefinite Programming & Convex Algebraic Geometry

Hilbert s 17th Problem to Semidefinite Programming & Convex Algebraic Geometry Hilbert s 17th Problem to Semidefinite Programming & Convex Algebraic Geometry Rekha R. Thomas University of Washington, Seattle References Monique Laurent, Sums of squares, moment matrices and optimization

More information

COURSE ON LMI PART I.2 GEOMETRY OF LMI SETS. Didier HENRION henrion

COURSE ON LMI PART I.2 GEOMETRY OF LMI SETS. Didier HENRION   henrion COURSE ON LMI PART I.2 GEOMETRY OF LMI SETS Didier HENRION www.laas.fr/ henrion October 2006 Geometry of LMI sets Given symmetric matrices F i we want to characterize the shape in R n of the LMI set F

More information

On approximations, complexity, and applications for copositive programming Gijben, Luuk

On approximations, complexity, and applications for copositive programming Gijben, Luuk University of Groningen On approximations, complexity, and applications for copositive programming Gijben, Luuk IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you

More information

Chapter 11. Approximation Algorithms. Slides by Kevin Wayne Pearson-Addison Wesley. All rights reserved.

Chapter 11. Approximation Algorithms. Slides by Kevin Wayne Pearson-Addison Wesley. All rights reserved. Chapter 11 Approximation Algorithms Slides by Kevin Wayne. Copyright @ 2005 Pearson-Addison Wesley. All rights reserved. 1 Approximation Algorithms Q. Suppose I need to solve an NP-hard problem. What should

More information

2-4 Zeros of Polynomial Functions

2-4 Zeros of Polynomial Functions Write a polynomial function of least degree with real coefficients in standard form that has the given zeros. 33. 2, 4, 3, 5 Using the Linear Factorization Theorem and the zeros 2, 4, 3, and 5, write f

More information

Completely Positive Reformulations for Polynomial Optimization

Completely Positive Reformulations for Polynomial Optimization manuscript No. (will be inserted by the editor) Completely Positive Reformulations for Polynomial Optimization Javier Peña Juan C. Vera Luis F. Zuluaga Received: date / Accepted: date Abstract Polynomial

More information

Finding normalized and modularity cuts by spectral clustering. Ljubjana 2010, October

Finding normalized and modularity cuts by spectral clustering. Ljubjana 2010, October Finding normalized and modularity cuts by spectral clustering Marianna Bolla Institute of Mathematics Budapest University of Technology and Economics marib@math.bme.hu Ljubjana 2010, October Outline Find

More information

CS675: Convex and Combinatorial Optimization Fall 2014 Combinatorial Problems as Linear Programs. Instructor: Shaddin Dughmi

CS675: Convex and Combinatorial Optimization Fall 2014 Combinatorial Problems as Linear Programs. Instructor: Shaddin Dughmi CS675: Convex and Combinatorial Optimization Fall 2014 Combinatorial Problems as Linear Programs Instructor: Shaddin Dughmi Outline 1 Introduction 2 Shortest Path 3 Algorithms for Single-Source Shortest

More information

Semi-Simultaneous Flows and Binary Constrained (Integer) Linear Programs

Semi-Simultaneous Flows and Binary Constrained (Integer) Linear Programs DEPARTMENT OF MATHEMATICAL SCIENCES Clemson University, South Carolina, USA Technical Report TR2006 07 EH Semi-Simultaneous Flows and Binary Constrained (Integer Linear Programs A. Engau and H. W. Hamacher

More information

Conic approach to quantum graph parameters using linear optimization over the completely positive semidefinite cone

Conic approach to quantum graph parameters using linear optimization over the completely positive semidefinite cone Conic approach to quantum graph parameters using linear optimization over the completely positive semidefinite cone Monique Laurent 1,2 and Teresa Piovesan 1 1 Centrum Wiskunde & Informatica (CWI), Amsterdam,

More information

Copositive and Semidefinite Relaxations of the Quadratic Assignment Problem (appeared in Discrete Optimization 6 (2009) )

Copositive and Semidefinite Relaxations of the Quadratic Assignment Problem (appeared in Discrete Optimization 6 (2009) ) Copositive and Semidefinite Relaxations of the Quadratic Assignment Problem (appeared in Discrete Optimization 6 (2009) 231 241) Janez Povh Franz Rendl July 22, 2009 Abstract Semidefinite relaxations of

More information

B-468 A Quadratically Constrained Quadratic Optimization Model for Completely Positive Cone Programming

B-468 A Quadratically Constrained Quadratic Optimization Model for Completely Positive Cone Programming B-468 A Quadratically Constrained Quadratic Optimization Model for Completely Positive Cone Programming Naohiko Arima, Sunyoung Kim and Masakazu Kojima September 2012 Abstract. We propose a class of quadratic

More information

Semidefinite Programming

Semidefinite Programming Semidefinite Programming Basics and SOS Fernando Mário de Oliveira Filho Campos do Jordão, 2 November 23 Available at: www.ime.usp.br/~fmario under talks Conic programming V is a real vector space h, i

More information

CS675: Convex and Combinatorial Optimization Fall 2016 Convex Optimization Problems. Instructor: Shaddin Dughmi

CS675: Convex and Combinatorial Optimization Fall 2016 Convex Optimization Problems. Instructor: Shaddin Dughmi CS675: Convex and Combinatorial Optimization Fall 2016 Convex Optimization Problems Instructor: Shaddin Dughmi Outline 1 Convex Optimization Basics 2 Common Classes 3 Interlude: Positive Semi-Definite

More information

A A x i x j i j (i, j) (j, i) Let. Compute the value of for and

A A x i x j i j (i, j) (j, i) Let. Compute the value of for and 7.2 - Quadratic Forms quadratic form on is a function defined on whose value at a vector in can be computed by an expression of the form, where is an symmetric matrix. The matrix R n Q R n x R n Q(x) =

More information

Matrix Mathematics. Theory, Facts, and Formulas with Application to Linear Systems Theory. Dennis S. Bernstein

Matrix Mathematics. Theory, Facts, and Formulas with Application to Linear Systems Theory. Dennis S. Bernstein Matrix Mathematics Theory, Facts, and Formulas with Application to Linear Systems Theory Dennis S. Bernstein PRINCETON UNIVERSITY PRESS PRINCETON AND OXFORD Contents Special Symbols xv Conventions, Notation,

More information

An alternative proof of the Barker, Berman, Plemmons (BBP) result on diagonal stability and extensions - Corrected Version

An alternative proof of the Barker, Berman, Plemmons (BBP) result on diagonal stability and extensions - Corrected Version 1 An alternative proof of the Barker, Berman, Plemmons (BBP) result on diagonal stability and extensions - Corrected Version Robert Shorten 1, Oliver Mason 1 and Christopher King 2 Abstract The original

More information

Lecture 19: Isometries, Positive operators, Polar and singular value decompositions; Unitary matrices and classical groups; Previews (1)

Lecture 19: Isometries, Positive operators, Polar and singular value decompositions; Unitary matrices and classical groups; Previews (1) Lecture 19: Isometries, Positive operators, Polar and singular value decompositions; Unitary matrices and classical groups; Previews (1) Travis Schedler Thurs, Nov 18, 2010 (version: Wed, Nov 17, 2:15

More information

MIT Algebraic techniques and semidefinite optimization May 9, Lecture 21. Lecturer: Pablo A. Parrilo Scribe:???

MIT Algebraic techniques and semidefinite optimization May 9, Lecture 21. Lecturer: Pablo A. Parrilo Scribe:??? MIT 6.972 Algebraic techniques and semidefinite optimization May 9, 2006 Lecture 2 Lecturer: Pablo A. Parrilo Scribe:??? In this lecture we study techniques to exploit the symmetry that can be present

More information

A Geometric Approach to Graph Isomorphism

A Geometric Approach to Graph Isomorphism A Geometric Approach to Graph Isomorphism Pawan Aurora and Shashank K Mehta Indian Institute of Technology, Kanpur - 208016, India {paurora,skmehta}@cse.iitk.ac.in Abstract. We present an integer linear

More information

Continuous Optimisation, Chpt 9: Semidefinite Problems

Continuous Optimisation, Chpt 9: Semidefinite Problems Continuous Optimisation, Chpt 9: Semidefinite Problems Peter J.C. Dickinson DMMP, University of Twente p.j.c.dickinson@utwente.nl http://dickinson.website/teaching/2016co.html version: 21/11/16 Monday

More information

CS675: Convex and Combinatorial Optimization Fall 2016 Combinatorial Problems as Linear and Convex Programs. Instructor: Shaddin Dughmi

CS675: Convex and Combinatorial Optimization Fall 2016 Combinatorial Problems as Linear and Convex Programs. Instructor: Shaddin Dughmi CS675: Convex and Combinatorial Optimization Fall 2016 Combinatorial Problems as Linear and Convex Programs Instructor: Shaddin Dughmi Outline 1 Introduction 2 Shortest Path 3 Algorithms for Single-Source

More information

THE LOVÁSZ THETA FUNCTION AND A SEMIDEFINITE PROGRAMMING RELAXATION OF VERTEX COVER

THE LOVÁSZ THETA FUNCTION AND A SEMIDEFINITE PROGRAMMING RELAXATION OF VERTEX COVER SIAM J. DISCRETE MATH. c 1998 Society for Industrial and Applied Mathematics Vol. 11, No., pp. 196 04, May 1998 00 THE LOVÁSZ THETA FUNCTION AND A SEMIDEFINITE PROGRAMMING RELAXATION OF VERTEX COVER JON

More information

Math 354 Transition graphs and subshifts November 26, 2014

Math 354 Transition graphs and subshifts November 26, 2014 Math 54 Transition graphs and subshifts November 6, 04. Transition graphs Let I be a closed interval in the real line. Suppose F : I I is function. Let I 0, I,..., I N be N closed subintervals in I with

More information

MIT Algebraic techniques and semidefinite optimization February 14, Lecture 3

MIT Algebraic techniques and semidefinite optimization February 14, Lecture 3 MI 6.97 Algebraic techniques and semidefinite optimization February 4, 6 Lecture 3 Lecturer: Pablo A. Parrilo Scribe: Pablo A. Parrilo In this lecture, we will discuss one of the most important applications

More information

7. Symmetric Matrices and Quadratic Forms

7. Symmetric Matrices and Quadratic Forms Linear Algebra 7. Symmetric Matrices and Quadratic Forms CSIE NCU 1 7. Symmetric Matrices and Quadratic Forms 7.1 Diagonalization of symmetric matrices 2 7.2 Quadratic forms.. 9 7.4 The singular value

More information

SCHUR IDEALS AND HOMOMORPHISMS OF THE SEMIDEFINITE CONE

SCHUR IDEALS AND HOMOMORPHISMS OF THE SEMIDEFINITE CONE SCHUR IDEALS AND HOMOMORPHISMS OF THE SEMIDEFINITE CONE BABHRU JOSHI AND M. SEETHARAMA GOWDA Abstract. We consider the semidefinite cone K n consisting of all n n real symmetric positive semidefinite matrices.

More information

On approximations, complexity, and applications for copositive programming Gijben, Luuk

On approximations, complexity, and applications for copositive programming Gijben, Luuk University of Groningen On approximations, complexity, and applications for copositive programming Gijben, Luuk IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you

More information

Chapter 3 Transformations

Chapter 3 Transformations Chapter 3 Transformations An Introduction to Optimization Spring, 2014 Wei-Ta Chu 1 Linear Transformations A function is called a linear transformation if 1. for every and 2. for every If we fix the bases

More information

Second Order Cone Programming Relaxation of Positive Semidefinite Constraint

Second Order Cone Programming Relaxation of Positive Semidefinite Constraint Research Reports on Mathematical and Computing Sciences Series B : Operations Research Department of Mathematical and Computing Sciences Tokyo Institute of Technology 2-12-1 Oh-Okayama, Meguro-ku, Tokyo

More information

A note on network reliability

A note on network reliability A note on network reliability Noga Alon Institute for Advanced Study, Princeton, NJ 08540 and Department of Mathematics Tel Aviv University, Tel Aviv, Israel Let G = (V, E) be a loopless undirected multigraph,

More information

Semidefinite Programming

Semidefinite Programming Semidefinite Programming Notes by Bernd Sturmfels for the lecture on June 26, 208, in the IMPRS Ringvorlesung Introduction to Nonlinear Algebra The transition from linear algebra to nonlinear algebra has

More information

Semidefinite Programming and Harmonic Analysis

Semidefinite Programming and Harmonic Analysis 1 / 74 Semidefinite Programming and Harmonic Analysis Cristóbal Guzmán CS 8803 - Discrete Fourier Analysis and Applications March 7, 2012 2 / 74 Motivation SDP gives best relaxations known for some combinatorial

More information