On duality gap in linear conic problems

Size: px
Start display at page:

Download "On duality gap in linear conic problems"

Transcription

1 On duality gap in linear conic problems C. Zălinescu Abstract In their paper Duality of linear conic problems A. Shapiro and A. Nemirovski considered two possible properties (A) and (B) for dual linear conic problems (P) and (D). The property (A) is If either (P) or (D) is feasible, then there is no duality gap between (P) and (D), while property (B) is If both (P) and (D) are feasible, then there is no duality gap between (P) and (D) and the optimal values val(p) and val(d) are finite. They showed that (A) holds if and only if the cone K is polyhedral, and gave some partial results related to (B). Later A. Shapiro conjectured that (B) holds if and only if all the nontrivial faces of the cone K are polyhedral. In this note we mainly prove that both the if and only if parts of this conjecture are not true by providing examples of closed convex cone in R 4 for which the corresponding implications are not valid. Moreover, we give alternative proofs for the results related to (B) established by A. Shapiro and A. Nemirovski. Key words: Linear conic problems, duality gap. 1 Introduction In [3] one considers the linear conic problem min x X c, x subject to Ax + b = 0 and x K and its dual max y Y b, y subject to A y + c K, where X and Y are finite dimensional vector spaces equipped with scalar products denoted,, A : X Y is a linear mapping and K X is a closed convex cone, while K := {x X x, x 0 x K} and A : Y X is the adjoint of A. In [3] it is raised the problem of characterizing those cones K for which property (B) holds, that is, there is no duality gap whenever both problems are feasible. Professor A. Shapiro conjectured later that property (B) holds if and only if all the nontrivial faces of the cone K are polyhedral. The main aim of this note is to show that this conjecture is not true. We also gave alternative proofs for some results in [3] related to property (B). We reformulate the above dual problems without using a linear operator A as in [1] (see also [5]). More precisely, consider X a finite dimensional normed vector space whose dual is denoted by X, K X a proper closed convex cone and M X a linear space. We denote x (x) by x, x for x X and x X and we consider the dual cone K + of K and the orthogonal space M of M : K + := {x X x, x 0 x K}, M := {x X x, x = 0 x M}. University Al. I. Cuza Iaşi, Faculty of Mathematics, Iaşi, Romania, zalinesc@uaic.ro. 1

2 Consider the linear conic problems (P) min x, x s.t. x x + M, x K, (D) max x, x s.t. x M, x x K +. Denoting x x by y, the problem (D) becomes max ( x, x x, y ) s.t. y x + M, y K +, or equivalently min x, y s.t. y x + M, y K +, that is, a problem of the same type as (P). We denote by F P (S P ) and F D (S D ) the feasible (solution) sets of problems (P) and (D), respectively; hence F P = (x + M) K, F D = M (x K + ). (1) If x is a feasible solution of (P), that is, x (x + M) K, and x is a feasible solution of (D) then x, x x, x, and so v P v D, where v P and v D are the values of the problems (P) and (D), respectively. Indeed, in such a case we have that x, x x, x = x, x x + x x, x = x, x x 0. Moreover, if x is a solution of (P), x is a solution of (D) and v P = v D then necessarily, x, x x = 0. (2) Conversely, if x is a feasible solution of (P), x is a feasible solution of (D) and x, x = x, x, then x is an optimal solution of (P) and x is an optimal solution of (D). If (P) has feasible solutions we may assume that x K. Indeed, take x (x + M) K. Then x + M = x + M, and so (x + M) K = ( x + M) K. Moreover, for x M we have that x, x = x x, x + x, x = x, x, which proves that the objective function of (D) remains unchanged on its feasible set. Similarly, we may assume that x K + if (D) is feasible. Because we are interested by the case in which both (P) and (D) are feasible, we assume that x K and x K +, and so x is feasible for (P) and 0 is feasible for (D); hence x, x v P v D 0. (3) Moreover, if (P) has optimal solutions then we may assume that x is even optimal for (P). Relation (3) shows that v P = v D if x, x = 0; in particular, this happens if x = 0 or x = 0. 2 Sufficient conditions for no duality gap Throughout this section we assume that X, K and M are as in the previous section and x K, x K +, x, x > 0. (4) The relation v P v D in (3) can be obtained also using perturbation functions (see [2], [4], as well as for other results and notation not explicitly mentioned in the sequel). Indeed, let us consider { x, x F : X Y R, F (x, y) := if x x + y + M, x K, (5) otherwise, 2

3 where Y := X. Clearly, F is a lsc proper convex function. Setting h(y) := inf x X F (x, y), we have that h(0) = v P 0. Moreover, F (x, y ) = sup { x, x + y, y F (x, y) x, y X} = sup { x, x + y, y x, x x K, y x x + M} = sup { x, x + x x + z, y x, x x K, z M} = x, y + sup { x, x + y x + z, y x K, z M} { x, y = if y M, x + y x K, otherwise, where K := K +. Hence the problem max ( F (0, y )) s.t. y Y becomes our problem (D). The next result will be useful in the sequel; possibly it is not new but we have not a reference for it. Proposition 1 Consider Y a normed space and F : X Y R a lsc proper convex function. Assume that 0 Pr Y (dom F ) and S := {x X F (x, 0) F (x, 0) x X} is nonempty and bounded. Then h : Y R, h(y) := inf x X F (x, y), is lsc at 0. Moreover, inf x X F (x, 0) = sup y Y ( F (0, y )) R. Proof. Take y n 0. Assume that lim inf h(y n ) < h(0). Then we may assume that h(y n ) < µ < h(0) for some µ R and every n. For n N there exists x n X with F (x n, y n ) < µ. If (x n ) has a bounded subsequence, because dim X <, we have that x nk x X for some subsequence (x nk ). Since F is lsc and y nk 0, we obtain that F (x, 0) lim inf F (x nk, y nk ) µ < h(0), a contradiction. Hence x n. We may assume that x n 1 x n u X \ {0}. Then (x n, y n, µ) epi F and x n 1 (x n, y n, µ) (u, 0, 0). Therefore, (u, 0, 0) (epi F ) = epi F, whence F (u, 0) 0. Taking x S, it follows that F (x + tu, 0) F (x, 0) + tf (u, 0) F (x, 0) for every t 0, which proves that x + tu S for every t 0, and so S is unbounded. This contradiction proves that h is lsc at 0. Since h(0) R (S being nonempty) and h is convex and lower semicontinuous at 0, we obtain (by [4, Th ]) that h(0) = h (0), and so inf x X F (x, 0) = sup y Y ( F (0, y )) R. If [(F P ) =] M K = {0} we have that F P is bounded (hence compact), and so the set S P is nonempty and compact; hence v P = v D by Proposition 1 applied to F defined in (5). Similarly, for M K + = {0} the set S D is nonempty and compact, and again v P = v D. In particular, if M is not proper we have that v P = v D. Assume that M K {0}, M K + {0}. (6) Suppose now that dim M = 1, that is, M = Ru for some u X \ {0}. From the preceding assumption we may assume that u K, and so x + R + u (x + M) K = F P x + Ru. If S P is not a singleton then u ker x := {x X x, x = 0}, and so x M. It follows that x is a feasible solution of (D), and so v P = x, x v D. Therefore, v P = v D. If S P is a singleton then clearly v P = v D (by Proposition 1). If dim M = dim X 1 then dim M = 1, and the conclusion follows changing the roles of (P) and (D). 3

4 We are interested now by the case in which (4) and (6) hold and, moreover, 2 dim M dim X 2; in particular, dim X 4. Clearly, from the definition of F in (5), Pr Y (dom F ) = K+M x. By a known fact (see e.g. [4, Th (viii)]) we have that inf x X F (x, 0) = max y Y ( F (0, y )) if 0 i Pr Y (dom F ), that is, x i (M + K) = M + i K; by i A we denote the algebraic interior (or intrinsic core) of the subset A of a real linear space E. Hence, in this case we have that v P = v D and (D) has optimal solutions whenever (D) is feasible (this is the case because x K + ). Because the dual problem of (D) is (P) and Pr X (dom F ) = x + K + M, we have that v P = v D and (P) has optimal solutions when x M + i (K + ) = i ( M + K +). Assume that x rbd(m +K) := cl(m +K)\ i (M +K), or, equivalently, (x + M) i K =. Then there exists u X such that x + v, u < x, u for all v M and x i K. Hence u M K + \ {0}. Since x K we obtain that x, u = 0, and so x + M ker u. It follows that (x + M) K = (x + M) (K ker u ). Setting K u := K ker u, this shows that the problem (P) is equivalent to where (PR) min x, x s.t. x x + M, x K u. Assume that K u is polyhedral. Then S P = S P R and v P = v P R = v DR v D 0, (DR) max x, x s.t. x M, x x (K u ) + = cl (K + + Ru ). If dim K u := dim(k u K u ) = 1 we get S P K u Rx. Because x is constant on S P and x, x > 0, we have necessarily that S P is a singleton. Therefore, v P = v D. The next result summerizes the above discussion. Proposition 2 Let x K and x K +. Then v P = v D if one of the following condition is satisfied: (a) dim M 1 or dim M 1; (b) x M + i K or x M + i (K + ); (c) dim K u = 1 for every u M K + \ {0}. If dim X 3 then (a) holds, and so we get [3, Prop. 4]. From (c) we get [3, Prop. 3] because in this case every nontrivial face of K has dimension one (see below the definition of a face of a convex set). The conclusion of Proposition 2 is well known when (b) holds. As mentioned by Prof. A. Shapiro in a discussion, the conjecture is that property (B) (in [3]) holds iff every face of cone K is polyhedral. In our framework, this conjecture translates as: 4

5 Conjecture 3 Let K X be a closed convex cone. Then v P = v D for all x K, x K + and all linear spaces M X if and only if all nontrivial faces of K are polyhedral. In [3] it is given an example of cone K R 4 for which the problems (P) and (D) have a finite duality gap; however, the cone K has a nontrivial face which is not polyhedral. Moreover, an example of nonpolyhedral cone K having only polyhedral nontrivial faces, one of which having dimension greater than one, for which Conjecture 3 is true is given in [3]. 3 Relations between the faces of cones and their bases Recall that a face of a convex set C E, where E is a real linear space, is a nonempty convex subset F C with the property: x, y C, λ (0, 1), λx + (1 λ)y F x, y F. Note that the convexity of F is essential. For example, taking C = [0, 1] R and F := {0, 1}, we have that x, y C, λ (0, 1), λx + (1 λ)y F x, y F ; however, F is not a face of C. The next two results are probably known; we give their proofs for readers convenience. They will be used in the next section. Proposition 4 Let P E be a convex cone having the base B E, that is, B is a nonempty convex subset of P such that any x P has a unique representation x = αu with α 0 and u B. Then F is a face of P iff F = {0} or there exists a face D of B such that F = R + D. Proof. Clearly, 0 / B and P is a pointed convex cone. Assume that D is a face of B and take x, y P with x y, and λ (0, 1) such that z := λx + (1 λ)y F = R + D. Then x = αa, y = βb, z = γc with α, β, γ 0, a, b B and c D. Because P is pointed, z 0, whence γ > 0 and α + β > 0. It follows that ( z = γc = (λα + (1 λ)β) and so γ = λα + (1 λ)β. Hence c = λα λα + (1 λ)β a + λα λα + (1 λ)β a + (1 λ)β λα + (1 λ)β b. ) (1 λ)β λα + (1 λ)β b, Since D is a face of B we get a, b D, whence x, y F. Hence F is a face of P. Let now {0} = F P be a face of P. Then F is a cone. Indeed, take z F \ {0} and α 0. If α > 1 then z = α 1 (αz) + (1 α 1 )0 with αz, 0 P. It follows that αz, 0 F ; in particular 0 = 0z F. If α (0, 1) there exists β > α and λ (0, 1) such that 1 = λα + (1 λ)β, and so z = λαz + (1 λ)βz. Since αz, βz P we get αz F. Hence F is a (convex) cone. Set D := {a B α > 0, x F : x = αa}; clearly D is nonempty because = F {0} and B is a base of P, and F = R + D because F is a cone. The set D is convex because for a, b D and λ (0, 1) we have a = α 1 x, b = β 1 y with α, β > 0 and x, y F, whence B λa + (1 λ)b = 1 (λα 1 x + (1 λ)β 1 y ), and so λa + (1 λ)b (taking into account the convexity of the cone F ). Moreover, assume that a, b B, λ (0, 1) are such that a b and λa + (1 λ)b D F. Since F is a face of P we obtain that a, b F, and so a, b D by the definition of D. Hence D is a face of B. Proposition 5 Let P E be a convex cone having the base B E. Then P is polyhedral iff P is algebraically closed and B is polyhedral. 5

6 Proof. It is well known that there exists φ 0 E, that is, φ 0 : E R is a linear functional, such that B = {x P φ 0 (x) = 1}. Assume first that P is polyhedral; clearly, P is algebraically closed (that is, the intersection of P with any line is closed in the line identified with R). Then there exist φ 1,..., φ k E such that P = {x E φ i (x) 0 i 1, k}. Then B = P {x E φ 0 (x) = 1}, and so B is polyhedral. Assume that B is polyhedral and P is algebraically closed; hence B = {x E φ i (x) γ i i 1, k} with φ i E and γ i R for i 1, k. Because B is nonempty, take x B; hence φ 0 (x) = 1 and φ i (x) γ i for i 1, k. Then P = {x E φ 0 (x) 0, φ i (x) γ i φ 0 (x) 0 i 0, k}. (7) Indeed, let x P \ {0}. Then α := φ 0 (x) > 0 and x := α 1 x B. It follows that γ i φ i (x ) = α 1 φ i (x), whence φ i (x) γ i φ 0 (x) 0 for every i 1, k. Hence the inclusion holds in (7). Take now x E with φ 0 (x) 0, φ i (x) γ i φ 0 (x) 0 for all i 0, k. Assume first that α := φ 0 (x) 0. Then α > 0 and setting x := α 1 x we obtain that φ 0 (x ) = 1 and φ i (x ) γ i φ 0 (x ) = φ i (x ) γ i 0 for all i 1, k. Hence x B, and so x R + B = P. Assume now that α = 0. Then, for λ > 0 we have φ 0 (x+λx) = λ > 0 and φ i (x+λx) γ i φ 0 (x+λx) = φ i (x) γ i φ 0 (x) + λ [φ i (x) γ i ] 0 for i 1, k. By the previous situation we get x + λx P. Since λ > 0 is arbitrary and P is algebraically closed we get x P. 4 Counter-examples to Conjecture 3 In the sequel we give examples showing that both implications in Conjecture 3 are not true. For constructing the examples we take into account the discussion in Section 2. We consider first the set A := A 1 A 2 R 3 with A 1 := { (x, y, z) R 3 x 2 + y 2 1, 0 z (1 x 2 y 2 ) 1/2}, A 2 := { (x, y, z) R 3 0 z (x 2 + y 2 ) 1/2 1 }. The sets A 1 and A 2 are compact convex sets with nonempty interior. Proposition 6 The set A is compact, convex and 0 int A. Moreover, all the nontrivial faces of A are polyhedral. Proof. The analytical proof of the statement is quite involved. However the picture in Figure 1 (a) is self-explanatory. Let us set K := R + (A {1}) R 4. (8) We have that K is a pointed closed convex cone with nonempty interior. Using Proposition 5 we obtain that K has only polyhedral nontrivial faces (of dimension 1 and 2). After some computation we obtain that where K + = R + (B {1}) R 4, (9) B := { (a, b, c) R 3 c [0, 1], a 2 + b 2 1 } { (a, b, c) R 3 c 0, a 2 + b 2 + c 2 1 } 6

7 z 0.0 z x 1 0 y y 0.0 x (a) The set A. (b) The set B. Figure 1: The sets A and B. (see Figure 1 (b)). We take x := (0, 1 2, 1 2, 1); because (0, 1 2, 1 2 ) A 2 we have that x K. We take also x := (0, 1, 1 2, 1); clearly x K +. Then x, x = 1 4 > 0. Let u := (0, 1, 1, 1) K + ; we have that x, u = 0. Then K u = K ker u = R + {(0, z + 1, z, 1) z [ 1, 0]} = {(0, α + β, α, β) β α 0}. Having in view the discussion in Section 2, we have to take M ker u with dim M = 2; so let M := {(α, β 2α, 2α, β) α, β R}. Then The feasible set for problem (P) is M = {(2α, β, α + β, β) α, β R}. F P = (x + M) K = (x + M) K u = { (0, β, 1 2, β ) β 0}. So x, x = 1 4 for every x F P := (x + M) K. Hence v P = 1 4 and the solution set of (P) is S P = F P. Let us evaluate v D. The feasible set of (D) is F D = M (x K + ). Take x F D ; hence x = (2α, β, α +β, β) with α, β R. Then ( 2α, β 1, 1 2 α β, 1 β) K+. Hence 1 β 1 2 α β 0 and 1 β [ 4α 2 + (β 1) 2] 1/2, in which case α = 0 and β 1 2, or, else 1 2 α β 0 and 1 β [ 4α 2 + (β 1) 2 + ( 1 2 α β)2] 1/2, in which case α = 0 and β = 1 2. Hence F D = {(0, β, β, β) R 4 β 1 2 } = (, 1 2 ] u. We have that x, x = 0 for every x F D, and so v D = 0 < 1 4 = v P. Note that (K u ) + = { (α, β, γ, δ) R 4 δ max{γ, β} }, and so F DR = {(2α, β, α + β, β) α, β R, α 1 2 }. It follows that v DR = sup{ 1 2 α α 1 2 } = 1 4 = v P = v P R. The previous example shows that the if part of Conjecture 3 is not true. Note that assuming that the Shapiro s conjecture was true we would have that all nontrivial faces of a closed cone K are polyhedral if and only if all nontrivial faces of K + are 7

8 polyhedral. The cone K defined by (8) has all the nontrivial faces polyhedral, while K + has a nontrivial face which is not polyhedral; this is F := R + {(x, y, 1, 1) x, y R, x 2 + y 2 1}. Having in view the previous example, one can ask if the following statement is true: Let K X be a closed convex cone. Then v P = v D for all x K, x K + and all linear spaces M X if and only if all nontrivial faces of K and K + are polyhedral. If this statement is true then one has the following consequence: If all nontrivial faces of K have dimension one than all nontrivial faces of K + are polyhedral. In the following example we show that this statement is not true. Consider D := { (x, y, z) R 3 N(x, y, z) 1 } (see Figure 2 (a)), where N(x, y, z) := x 2 + y 2 + z 2 + x 2 + y 2. Clearly N is a norm on R 3 which is strictly convex. It follows that D is strictly convex, and so any face of D is a singleton. Hence all the nontrivial faces of the cone K D := R + (D {1}) R 4 have dimension 1. It follows that K + D = R +(D {1}) with D = {u R 3 N (u) 1}, where N is the dual norm of N. But { c if a 2 + b 2 c, N (a, b, c) = sup {ax + by + cz N(x, y, z) 1} = a 2 +b 2 +c 2 2 if a 2 + b 2 > c. a 2 +b 2 Hence (see Figure 2 (b)) D = { (a, b, c) a 2 + b 2 1, c 1 } {(a, b, c) 1 a 2 + b c 2 }. One observes that F := {(a, b, 1) a 2 + b 2 1} is a non polyhedral face of D, and so K + has non polyhedral nontrivial faces. z 0.0 z y x y x 1 2 (a) The set D. (b) The set D. Figure 2: The sets D and D. Using Proposition 2 we obtain that property (B) holds for the cone K in the previous example, and so property (B) also holds for the cone K replaced by the cone K +. This shows that the only if part of Conjecture 3 is not true. As mentioned also by Prof. A. Shapiro, one can formulate the following problem. Problem 7 Let K X be a closed convex cone such that all nontrivial faces of K and K + are polyhedral. Is it true that v P = v D for all x K, x K + and all linear spaces M X? 8

9 Acknowledgement. I thank Prof. A. Shapiro for discussions and his remarks on the present paper. References [1] Y. E. Nesterov, A. S. Nemirovski, Interior point Polynomial Algorithms in Convex Programming: Theory and Applications. Philadelphia: Society for Industrial and Applied Mathematics, [2] R. T. Rockafellar, Convex Analysis, Princeton Univ. Press, Princeton, N.J., [3] A. Shapiro, A. S. Nemirovski, Duality of linear conic problems, Preprint, School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, GA 2003 ( HTML/2003/12/793.html). [4] C. Zălinescu, Convex Analysis in General Vector Spaces, World Scientific, Singapore, [5] C. Zălinescu, On zero duality gap and the Farkas lemma for conic programming, Math. Oper. Res. 33 (2008),

Some Properties of the Augmented Lagrangian in Cone Constrained Optimization

Some Properties of the Augmented Lagrangian in Cone Constrained Optimization MATHEMATICS OF OPERATIONS RESEARCH Vol. 29, No. 3, August 2004, pp. 479 491 issn 0364-765X eissn 1526-5471 04 2903 0479 informs doi 10.1287/moor.1040.0103 2004 INFORMS Some Properties of the Augmented

More information

On duality theory of conic linear problems

On duality theory of conic linear problems On duality theory of conic linear problems Alexander Shapiro School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia 3332-25, USA e-mail: ashapiro@isye.gatech.edu

More information

On Semicontinuity of Convex-valued Multifunctions and Cesari s Property (Q)

On Semicontinuity of Convex-valued Multifunctions and Cesari s Property (Q) On Semicontinuity of Convex-valued Multifunctions and Cesari s Property (Q) Andreas Löhne May 2, 2005 (last update: November 22, 2005) Abstract We investigate two types of semicontinuity for set-valued

More information

The local equicontinuity of a maximal monotone operator

The local equicontinuity of a maximal monotone operator arxiv:1410.3328v2 [math.fa] 3 Nov 2014 The local equicontinuity of a maximal monotone operator M.D. Voisei Abstract The local equicontinuity of an operator T : X X with proper Fitzpatrick function ϕ T

More information

On the use of semi-closed sets and functions in convex analysis

On the use of semi-closed sets and functions in convex analysis Open Math. 2015; 13: 1 5 Open Mathematics Open Access Research Article Constantin Zălinescu* On the use of semi-closed sets and functions in convex analysis Abstract: The main aim of this short note is

More information

Week 3: Faces of convex sets

Week 3: Faces of convex sets Week 3: Faces of convex sets Conic Optimisation MATH515 Semester 018 Vera Roshchina School of Mathematics and Statistics, UNSW August 9, 018 Contents 1. Faces of convex sets 1. Minkowski theorem 3 3. Minimal

More information

Quasi-relative interior and optimization

Quasi-relative interior and optimization Quasi-relative interior and optimization Constantin Zălinescu University Al. I. Cuza Iaşi, Faculty of Mathematics CRESWICK, 2016 1 Aim and framework Aim Framework and notation The quasi-relative interior

More information

Lecture 5. The Dual Cone and Dual Problem

Lecture 5. The Dual Cone and Dual Problem IE 8534 1 Lecture 5. The Dual Cone and Dual Problem IE 8534 2 For a convex cone K, its dual cone is defined as K = {y x, y 0, x K}. The inner-product can be replaced by x T y if the coordinates of the

More information

Local strong convexity and local Lipschitz continuity of the gradient of convex functions

Local strong convexity and local Lipschitz continuity of the gradient of convex functions Local strong convexity and local Lipschitz continuity of the gradient of convex functions R. Goebel and R.T. Rockafellar May 23, 2007 Abstract. Given a pair of convex conjugate functions f and f, we investigate

More information

Summer School: Semidefinite Optimization

Summer School: Semidefinite Optimization Summer School: Semidefinite Optimization Christine Bachoc Université Bordeaux I, IMB Research Training Group Experimental and Constructive Algebra Haus Karrenberg, Sept. 3 - Sept. 7, 2012 Duality Theory

More information

ON GENERALIZED-CONVEX CONSTRAINED MULTI-OBJECTIVE OPTIMIZATION

ON GENERALIZED-CONVEX CONSTRAINED MULTI-OBJECTIVE OPTIMIZATION ON GENERALIZED-CONVEX CONSTRAINED MULTI-OBJECTIVE OPTIMIZATION CHRISTIAN GÜNTHER AND CHRISTIANE TAMMER Abstract. In this paper, we consider multi-objective optimization problems involving not necessarily

More information

ON A CLASS OF NONSMOOTH COMPOSITE FUNCTIONS

ON A CLASS OF NONSMOOTH COMPOSITE FUNCTIONS MATHEMATICS OF OPERATIONS RESEARCH Vol. 28, No. 4, November 2003, pp. 677 692 Printed in U.S.A. ON A CLASS OF NONSMOOTH COMPOSITE FUNCTIONS ALEXANDER SHAPIRO We discuss in this paper a class of nonsmooth

More information

Lecture 9 Monotone VIs/CPs Properties of cones and some existence results. October 6, 2008

Lecture 9 Monotone VIs/CPs Properties of cones and some existence results. October 6, 2008 Lecture 9 Monotone VIs/CPs Properties of cones and some existence results October 6, 2008 Outline Properties of cones Existence results for monotone CPs/VIs Polyhedrality of solution sets Game theory:

More information

4. Algebra and Duality

4. Algebra and Duality 4-1 Algebra and Duality P. Parrilo and S. Lall, CDC 2003 2003.12.07.01 4. Algebra and Duality Example: non-convex polynomial optimization Weak duality and duality gap The dual is not intrinsic The cone

More information

On smoothness properties of optimal value functions at the boundary of their domain under complete convexity

On smoothness properties of optimal value functions at the boundary of their domain under complete convexity On smoothness properties of optimal value functions at the boundary of their domain under complete convexity Oliver Stein # Nathan Sudermann-Merx June 14, 2013 Abstract This article studies continuity

More information

Preprint February 19, 2018 (1st version October 31, 2017) Pareto efficient solutions in multi-objective optimization involving forbidden regions 1

Preprint February 19, 2018 (1st version October 31, 2017) Pareto efficient solutions in multi-objective optimization involving forbidden regions 1 Preprint February 19, 2018 1st version October 31, 2017) Pareto efficient solutions in multi-objective optimization involving forbidden regions 1 by CHRISTIAN GÜNTHER Martin Luther University Halle-Wittenberg

More information

Convex Optimization M2

Convex Optimization M2 Convex Optimization M2 Lecture 3 A. d Aspremont. Convex Optimization M2. 1/49 Duality A. d Aspremont. Convex Optimization M2. 2/49 DMs DM par email: dm.daspremont@gmail.com A. d Aspremont. Convex Optimization

More information

Chapter 1. Preliminaries

Chapter 1. Preliminaries Introduction This dissertation is a reading of chapter 4 in part I of the book : Integer and Combinatorial Optimization by George L. Nemhauser & Laurence A. Wolsey. The chapter elaborates links between

More information

Continuous Sets and Non-Attaining Functionals in Reflexive Banach Spaces

Continuous Sets and Non-Attaining Functionals in Reflexive Banach Spaces Laboratoire d Arithmétique, Calcul formel et d Optimisation UMR CNRS 6090 Continuous Sets and Non-Attaining Functionals in Reflexive Banach Spaces Emil Ernst Michel Théra Rapport de recherche n 2004-04

More information

Example: feasibility. Interpretation as formal proof. Example: linear inequalities and Farkas lemma

Example: feasibility. Interpretation as formal proof. Example: linear inequalities and Farkas lemma 4-1 Algebra and Duality P. Parrilo and S. Lall 2006.06.07.01 4. Algebra and Duality Example: non-convex polynomial optimization Weak duality and duality gap The dual is not intrinsic The cone of valid

More information

POLARS AND DUAL CONES

POLARS AND DUAL CONES POLARS AND DUAL CONES VERA ROSHCHINA Abstract. The goal of this note is to remind the basic definitions of convex sets and their polars. For more details see the classic references [1, 2] and [3] for polytopes.

More information

Semi-infinite programming, duality, discretization and optimality conditions

Semi-infinite programming, duality, discretization and optimality conditions Semi-infinite programming, duality, discretization and optimality conditions Alexander Shapiro School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0205,

More information

ZERO DUALITY GAP FOR CONVEX PROGRAMS: A GENERAL RESULT

ZERO DUALITY GAP FOR CONVEX PROGRAMS: A GENERAL RESULT ZERO DUALITY GAP FOR CONVEX PROGRAMS: A GENERAL RESULT EMIL ERNST AND MICHEL VOLLE Abstract. This article addresses a general criterion providing a zero duality gap for convex programs in the setting of

More information

The Subdifferential of Convex Deviation Measures and Risk Functions

The Subdifferential of Convex Deviation Measures and Risk Functions The Subdifferential of Convex Deviation Measures and Risk Functions Nicole Lorenz Gert Wanka In this paper we give subdifferential formulas of some convex deviation measures using their conjugate functions

More information

LECTURE 25: REVIEW/EPILOGUE LECTURE OUTLINE

LECTURE 25: REVIEW/EPILOGUE LECTURE OUTLINE LECTURE 25: REVIEW/EPILOGUE LECTURE OUTLINE CONVEX ANALYSIS AND DUALITY Basic concepts of convex analysis Basic concepts of convex optimization Geometric duality framework - MC/MC Constrained optimization

More information

Appendix B Convex analysis

Appendix B Convex analysis This version: 28/02/2014 Appendix B Convex analysis In this appendix we review a few basic notions of convexity and related notions that will be important for us at various times. B.1 The Hausdorff distance

More information

When are Sums Closed?

When are Sums Closed? Division of the Humanities and Social Sciences Ec 181 KC Border Convex Analysis and Economic Theory Fall 2018 Winter 2019 Topic 20: When are Sums Closed? 20.1 Is a sum of closed sets closed? Example 0.2.2

More information

Constraint qualifications for convex inequality systems with applications in constrained optimization

Constraint qualifications for convex inequality systems with applications in constrained optimization Constraint qualifications for convex inequality systems with applications in constrained optimization Chong Li, K. F. Ng and T. K. Pong Abstract. For an inequality system defined by an infinite family

More information

A Geometric Framework for Nonconvex Optimization Duality using Augmented Lagrangian Functions

A Geometric Framework for Nonconvex Optimization Duality using Augmented Lagrangian Functions A Geometric Framework for Nonconvex Optimization Duality using Augmented Lagrangian Functions Angelia Nedić and Asuman Ozdaglar April 15, 2006 Abstract We provide a unifying geometric framework for the

More information

Optimality Conditions for Nonsmooth Convex Optimization

Optimality Conditions for Nonsmooth Convex Optimization Optimality Conditions for Nonsmooth Convex Optimization Sangkyun Lee Oct 22, 2014 Let us consider a convex function f : R n R, where R is the extended real field, R := R {, + }, which is proper (f never

More information

Math 341: Convex Geometry. Xi Chen

Math 341: Convex Geometry. Xi Chen Math 341: Convex Geometry Xi Chen 479 Central Academic Building, University of Alberta, Edmonton, Alberta T6G 2G1, CANADA E-mail address: xichen@math.ualberta.ca CHAPTER 1 Basics 1. Euclidean Geometry

More information

Convex series of convex functions with applications to Statistical Mechanics

Convex series of convex functions with applications to Statistical Mechanics Convex series of convex functions with applications to Statistical Mechanics Constantin Zălinescu University Alexandru Ioan Cuza Iaşi Faculty of Mathematics zalinesc@uaic.ro Melbourne, MODU2016 Motivation

More information

Chapter 2. Convex Sets: basic results

Chapter 2. Convex Sets: basic results Chapter 2 Convex Sets: basic results In this chapter, we introduce one of the most important tools in the mathematical approach to Economics, namely the theory of convex sets. Almost every situation we

More information

On Total Convexity, Bregman Projections and Stability in Banach Spaces

On Total Convexity, Bregman Projections and Stability in Banach Spaces Journal of Convex Analysis Volume 11 (2004), No. 1, 1 16 On Total Convexity, Bregman Projections and Stability in Banach Spaces Elena Resmerita Department of Mathematics, University of Haifa, 31905 Haifa,

More information

Optimization and Optimal Control in Banach Spaces

Optimization and Optimal Control in Banach Spaces Optimization and Optimal Control in Banach Spaces Bernhard Schmitzer October 19, 2017 1 Convex non-smooth optimization with proximal operators Remark 1.1 (Motivation). Convex optimization: easier to solve,

More information

IE 521 Convex Optimization Homework #1 Solution

IE 521 Convex Optimization Homework #1 Solution IE 521 Convex Optimization Homework #1 Solution your NAME here your NetID here February 13, 2019 Instructions. Homework is due Wednesday, February 6, at 1:00pm; no late homework accepted. Please use the

More information

USING FUNCTIONAL ANALYSIS AND SOBOLEV SPACES TO SOLVE POISSON S EQUATION

USING FUNCTIONAL ANALYSIS AND SOBOLEV SPACES TO SOLVE POISSON S EQUATION USING FUNCTIONAL ANALYSIS AND SOBOLEV SPACES TO SOLVE POISSON S EQUATION YI WANG Abstract. We study Banach and Hilbert spaces with an eye towards defining weak solutions to elliptic PDE. Using Lax-Milgram

More information

(convex combination!). Use convexity of f and multiply by the common denominator to get. Interchanging the role of x and y, we obtain that f is ( 2M ε

(convex combination!). Use convexity of f and multiply by the common denominator to get. Interchanging the role of x and y, we obtain that f is ( 2M ε 1. Continuity of convex functions in normed spaces In this chapter, we consider continuity properties of real-valued convex functions defined on open convex sets in normed spaces. Recall that every infinitedimensional

More information

Convex Analysis and Optimization Chapter 4 Solutions

Convex Analysis and Optimization Chapter 4 Solutions Convex Analysis and Optimization Chapter 4 Solutions Dimitri P. Bertsekas with Angelia Nedić and Asuman E. Ozdaglar Massachusetts Institute of Technology Athena Scientific, Belmont, Massachusetts http://www.athenasc.com

More information

Maximal Monotone Inclusions and Fitzpatrick Functions

Maximal Monotone Inclusions and Fitzpatrick Functions JOTA manuscript No. (will be inserted by the editor) Maximal Monotone Inclusions and Fitzpatrick Functions J. M. Borwein J. Dutta Communicated by Michel Thera. Abstract In this paper, we study maximal

More information

CONSTRAINT QUALIFICATIONS, LAGRANGIAN DUALITY & SADDLE POINT OPTIMALITY CONDITIONS

CONSTRAINT QUALIFICATIONS, LAGRANGIAN DUALITY & SADDLE POINT OPTIMALITY CONDITIONS CONSTRAINT QUALIFICATIONS, LAGRANGIAN DUALITY & SADDLE POINT OPTIMALITY CONDITIONS A Dissertation Submitted For The Award of the Degree of Master of Philosophy in Mathematics Neelam Patel School of Mathematics

More information

REAL RENORMINGS ON COMPLEX BANACH SPACES

REAL RENORMINGS ON COMPLEX BANACH SPACES REAL RENORMINGS ON COMPLEX BANACH SPACES F. J. GARCÍA PACHECO AND A. MIRALLES Abstract. In this paper we provide two ways of obtaining real Banach spaces that cannot come from complex spaces. In concrete

More information

Continuity of convex functions in normed spaces

Continuity of convex functions in normed spaces Continuity of convex functions in normed spaces In this chapter, we consider continuity properties of real-valued convex functions defined on open convex sets in normed spaces. Recall that every infinitedimensional

More information

Monotone operators and bigger conjugate functions

Monotone operators and bigger conjugate functions Monotone operators and bigger conjugate functions Heinz H. Bauschke, Jonathan M. Borwein, Xianfu Wang, and Liangjin Yao August 12, 2011 Abstract We study a question posed by Stephen Simons in his 2008

More information

Division of the Humanities and Social Sciences. Sums of sets, etc.

Division of the Humanities and Social Sciences. Sums of sets, etc. Division of the Humanities and Social Sciences Sums of sets, etc. KC Border September 2002 Rev. November 2012 Rev. September 2013 If E and F are subsets of R m, define the sum E + F = {x + y : x E; y F

More information

Lecture 5. Theorems of Alternatives and Self-Dual Embedding

Lecture 5. Theorems of Alternatives and Self-Dual Embedding IE 8534 1 Lecture 5. Theorems of Alternatives and Self-Dual Embedding IE 8534 2 A system of linear equations may not have a solution. It is well known that either Ax = c has a solution, or A T y = 0, c

More information

A Parametric Simplex Algorithm for Linear Vector Optimization Problems

A Parametric Simplex Algorithm for Linear Vector Optimization Problems A Parametric Simplex Algorithm for Linear Vector Optimization Problems Birgit Rudloff Firdevs Ulus Robert Vanderbei July 9, 2015 Abstract In this paper, a parametric simplex algorithm for solving linear

More information

Sequential Pareto Subdifferential Sum Rule And Sequential Effi ciency

Sequential Pareto Subdifferential Sum Rule And Sequential Effi ciency Applied Mathematics E-Notes, 16(2016), 133-143 c ISSN 1607-2510 Available free at mirror sites of http://www.math.nthu.edu.tw/ amen/ Sequential Pareto Subdifferential Sum Rule And Sequential Effi ciency

More information

Convex Optimization Notes

Convex Optimization Notes Convex Optimization Notes Jonathan Siegel January 2017 1 Convex Analysis This section is devoted to the study of convex functions f : B R {+ } and convex sets U B, for B a Banach space. The case of B =

More information

On the projection onto a finitely generated cone

On the projection onto a finitely generated cone Acta Cybernetica 00 (0000) 1 15. On the projection onto a finitely generated cone Miklós Ujvári Abstract In the paper we study the properties of the projection onto a finitely generated cone. We show for

More information

Dedicated to Michel Théra in honor of his 70th birthday

Dedicated to Michel Théra in honor of his 70th birthday VARIATIONAL GEOMETRIC APPROACH TO GENERALIZED DIFFERENTIAL AND CONJUGATE CALCULI IN CONVEX ANALYSIS B. S. MORDUKHOVICH 1, N. M. NAM 2, R. B. RECTOR 3 and T. TRAN 4. Dedicated to Michel Théra in honor of

More information

A Dual Condition for the Convex Subdifferential Sum Formula with Applications

A Dual Condition for the Convex Subdifferential Sum Formula with Applications Journal of Convex Analysis Volume 12 (2005), No. 2, 279 290 A Dual Condition for the Convex Subdifferential Sum Formula with Applications R. S. Burachik Engenharia de Sistemas e Computacao, COPPE-UFRJ

More information

Translative Sets and Functions and their Applications to Risk Measure Theory and Nonlinear Separation

Translative Sets and Functions and their Applications to Risk Measure Theory and Nonlinear Separation Translative Sets and Functions and their Applications to Risk Measure Theory and Nonlinear Separation Andreas H. Hamel Abstract Recently defined concepts such as nonlinear separation functionals due to

More information

Appendix A: Separation theorems in IR n

Appendix A: Separation theorems in IR n Appendix A: Separation theorems in IR n These notes provide a number of separation theorems for convex sets in IR n. We start with a basic result, give a proof with the help on an auxiliary result and

More information

Stability in linear optimization under perturbations of the left-hand side coefficients 1

Stability in linear optimization under perturbations of the left-hand side coefficients 1 Stability in linear optimization under perturbations of the left-hand side coefficients 1 A. Daniilidis, M.-A. Goberna, M.A. Lopez, R. Lucchetti Abstract. This paper studies stability properties of linear

More information

GEORGIA INSTITUTE OF TECHNOLOGY H. MILTON STEWART SCHOOL OF INDUSTRIAL AND SYSTEMS ENGINEERING LECTURE NOTES OPTIMIZATION III

GEORGIA INSTITUTE OF TECHNOLOGY H. MILTON STEWART SCHOOL OF INDUSTRIAL AND SYSTEMS ENGINEERING LECTURE NOTES OPTIMIZATION III GEORGIA INSTITUTE OF TECHNOLOGY H. MILTON STEWART SCHOOL OF INDUSTRIAL AND SYSTEMS ENGINEERING LECTURE NOTES OPTIMIZATION III CONVEX ANALYSIS NONLINEAR PROGRAMMING THEORY NONLINEAR PROGRAMMING ALGORITHMS

More information

SOME STABILITY RESULTS FOR THE SEMI-AFFINE VARIATIONAL INEQUALITY PROBLEM. 1. Introduction

SOME STABILITY RESULTS FOR THE SEMI-AFFINE VARIATIONAL INEQUALITY PROBLEM. 1. Introduction ACTA MATHEMATICA VIETNAMICA 271 Volume 29, Number 3, 2004, pp. 271-280 SOME STABILITY RESULTS FOR THE SEMI-AFFINE VARIATIONAL INEQUALITY PROBLEM NGUYEN NANG TAM Abstract. This paper establishes two theorems

More information

FENCHEL DUALITY, FITZPATRICK FUNCTIONS AND MAXIMAL MONOTONICITY S. SIMONS AND C. ZĂLINESCU

FENCHEL DUALITY, FITZPATRICK FUNCTIONS AND MAXIMAL MONOTONICITY S. SIMONS AND C. ZĂLINESCU FENCHEL DUALITY, FITZPATRICK FUNCTIONS AND MAXIMAL MONOTONICITY S. SIMONS AND C. ZĂLINESCU This paper is dedicated to Simon Fitzpatrick, in recognition of his amazing insights ABSTRACT. We show in this

More information

Semicontinuous functions and convexity

Semicontinuous functions and convexity Semicontinuous functions and convexity Jordan Bell jordan.bell@gmail.com Department of Mathematics, University of Toronto April 3, 2014 1 Lattices If (A, ) is a partially ordered set and S is a subset

More information

Some Properties of Convex Hulls of Integer Points Contained in General Convex Sets

Some Properties of Convex Hulls of Integer Points Contained in General Convex Sets Some Properties of Convex Hulls of Integer Points Contained in General Convex Sets Santanu S. Dey and Diego A. Morán R. H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute

More information

A Note on Nonconvex Minimax Theorem with Separable Homogeneous Polynomials

A Note on Nonconvex Minimax Theorem with Separable Homogeneous Polynomials A Note on Nonconvex Minimax Theorem with Separable Homogeneous Polynomials G. Y. Li Communicated by Harold P. Benson Abstract The minimax theorem for a convex-concave bifunction is a fundamental theorem

More information

Chapter 2 Convex Analysis

Chapter 2 Convex Analysis Chapter 2 Convex Analysis The theory of nonsmooth analysis is based on convex analysis. Thus, we start this chapter by giving basic concepts and results of convexity (for further readings see also [202,

More information

Robust Farkas Lemma for Uncertain Linear Systems with Applications

Robust Farkas Lemma for Uncertain Linear Systems with Applications Robust Farkas Lemma for Uncertain Linear Systems with Applications V. Jeyakumar and G. Li Revised Version: July 8, 2010 Abstract We present a robust Farkas lemma, which provides a new generalization of

More information

The sum of two maximal monotone operator is of type FPV

The sum of two maximal monotone operator is of type FPV CJMS. 5(1)(2016), 17-21 Caspian Journal of Mathematical Sciences (CJMS) University of Mazandaran, Iran http://cjms.journals.umz.ac.ir ISSN: 1735-0611 The sum of two maximal monotone operator is of type

More information

A New Fenchel Dual Problem in Vector Optimization

A New Fenchel Dual Problem in Vector Optimization A New Fenchel Dual Problem in Vector Optimization Radu Ioan Boţ Anca Dumitru Gert Wanka Abstract We introduce a new Fenchel dual for vector optimization problems inspired by the form of the Fenchel dual

More information

A Unified Analysis of Nonconvex Optimization Duality and Penalty Methods with General Augmenting Functions

A Unified Analysis of Nonconvex Optimization Duality and Penalty Methods with General Augmenting Functions A Unified Analysis of Nonconvex Optimization Duality and Penalty Methods with General Augmenting Functions Angelia Nedić and Asuman Ozdaglar April 16, 2006 Abstract In this paper, we study a unifying framework

More information

A comparison of alternative c-conjugate dual problems in innite convex optimization

A comparison of alternative c-conjugate dual problems in innite convex optimization A comparison of alternative c-conjugate dual problems in innite convex optimization M.D. Fajardo 1, J. Vidal Department of Mathematics University of Alicante, 03080 Alicante, Spain Abstract In this work

More information

6 Lecture 6: More constructions with Huber rings

6 Lecture 6: More constructions with Huber rings 6 Lecture 6: More constructions with Huber rings 6.1 Introduction Recall from Definition 5.2.4 that a Huber ring is a commutative topological ring A equipped with an open subring A 0, such that the subspace

More information

Convex Optimization Theory. Chapter 5 Exercises and Solutions: Extended Version

Convex Optimization Theory. Chapter 5 Exercises and Solutions: Extended Version Convex Optimization Theory Chapter 5 Exercises and Solutions: Extended Version Dimitri P. Bertsekas Massachusetts Institute of Technology Athena Scientific, Belmont, Massachusetts http://www.athenasc.com

More information

Strong Dual for Conic Mixed-Integer Programs

Strong Dual for Conic Mixed-Integer Programs Strong Dual for Conic Mixed-Integer Programs Diego A. Morán R. Santanu S. Dey Juan Pablo Vielma July 14, 011 Abstract Mixed-integer conic programming is a generalization of mixed-integer linear programming.

More information

CONVEX OPTIMIZATION VIA LINEARIZATION. Miguel A. Goberna. Universidad de Alicante. Iberian Conference on Optimization Coimbra, November, 2006

CONVEX OPTIMIZATION VIA LINEARIZATION. Miguel A. Goberna. Universidad de Alicante. Iberian Conference on Optimization Coimbra, November, 2006 CONVEX OPTIMIZATION VIA LINEARIZATION Miguel A. Goberna Universidad de Alicante Iberian Conference on Optimization Coimbra, 16-18 November, 2006 Notation X denotes a l.c. Hausdorff t.v.s and X its topological

More information

The proximal mapping

The proximal mapping The proximal mapping http://bicmr.pku.edu.cn/~wenzw/opt-2016-fall.html Acknowledgement: this slides is based on Prof. Lieven Vandenberghes lecture notes Outline 2/37 1 closed function 2 Conjugate function

More information

The small ball property in Banach spaces (quantitative results)

The small ball property in Banach spaces (quantitative results) The small ball property in Banach spaces (quantitative results) Ehrhard Behrends Abstract A metric space (M, d) is said to have the small ball property (sbp) if for every ε 0 > 0 there exists a sequence

More information

SOME REMARKS ON SUBDIFFERENTIABILITY OF CONVEX FUNCTIONS

SOME REMARKS ON SUBDIFFERENTIABILITY OF CONVEX FUNCTIONS Applied Mathematics E-Notes, 5(2005), 150-156 c ISSN 1607-2510 Available free at mirror sites of http://www.math.nthu.edu.tw/ amen/ SOME REMARKS ON SUBDIFFERENTIABILITY OF CONVEX FUNCTIONS Mohamed Laghdir

More information

GEOMETRIC APPROACH TO CONVEX SUBDIFFERENTIAL CALCULUS October 10, Dedicated to Franco Giannessi and Diethard Pallaschke with great respect

GEOMETRIC APPROACH TO CONVEX SUBDIFFERENTIAL CALCULUS October 10, Dedicated to Franco Giannessi and Diethard Pallaschke with great respect GEOMETRIC APPROACH TO CONVEX SUBDIFFERENTIAL CALCULUS October 10, 2018 BORIS S. MORDUKHOVICH 1 and NGUYEN MAU NAM 2 Dedicated to Franco Giannessi and Diethard Pallaschke with great respect Abstract. In

More information

Course Notes for EE227C (Spring 2018): Convex Optimization and Approximation

Course Notes for EE227C (Spring 2018): Convex Optimization and Approximation Course Notes for EE227C (Spring 2018): Convex Optimization and Approximation Instructor: Moritz Hardt Email: hardt+ee227c@berkeley.edu Graduate Instructor: Max Simchowitz Email: msimchow+ee227c@berkeley.edu

More information

CONTINUOUS CONVEX SETS AND ZERO DUALITY GAP FOR CONVEX PROGRAMS

CONTINUOUS CONVEX SETS AND ZERO DUALITY GAP FOR CONVEX PROGRAMS CONTINUOUS CONVEX SETS AND ZERO DUALITY GAP FOR CONVEX PROGRAMS EMIL ERNST AND MICHEL VOLLE ABSTRACT. This article uses classical notions of convex analysis over euclidean spaces, like Gale & Klee s boundary

More information

LECTURE 12 LECTURE OUTLINE. Subgradients Fenchel inequality Sensitivity in constrained optimization Subdifferential calculus Optimality conditions

LECTURE 12 LECTURE OUTLINE. Subgradients Fenchel inequality Sensitivity in constrained optimization Subdifferential calculus Optimality conditions LECTURE 12 LECTURE OUTLINE Subgradients Fenchel inequality Sensitivity in constrained optimization Subdifferential calculus Optimality conditions Reading: Section 5.4 All figures are courtesy of Athena

More information

Refined optimality conditions for differences of convex functions

Refined optimality conditions for differences of convex functions Noname manuscript No. (will be inserted by the editor) Refined optimality conditions for differences of convex functions Tuomo Valkonen the date of receipt and acceptance should be inserted later Abstract

More information

Self-equilibrated Functions in Dual Vector Spaces: a Boundedness Criterion

Self-equilibrated Functions in Dual Vector Spaces: a Boundedness Criterion Self-equilibrated Functions in Dual Vector Spaces: a Boundedness Criterion Michel Théra LACO, UMR-CNRS 6090, Université de Limoges michel.thera@unilim.fr reporting joint work with E. Ernst and M. Volle

More information

Largest dual ellipsoids inscribed in dual cones

Largest dual ellipsoids inscribed in dual cones Largest dual ellipsoids inscribed in dual cones M. J. Todd June 23, 2005 Abstract Suppose x and s lie in the interiors of a cone K and its dual K respectively. We seek dual ellipsoidal norms such that

More information

Asteroide Santana, Santanu S. Dey. December 4, School of Industrial and Systems Engineering, Georgia Institute of Technology

Asteroide Santana, Santanu S. Dey. December 4, School of Industrial and Systems Engineering, Georgia Institute of Technology for Some for Asteroide Santana, Santanu S. Dey School of Industrial Systems Engineering, Georgia Institute of Technology December 4, 2016 1 / 38 1 1.1 Conic integer programs for Conic integer programs

More information

SEMIDEFINITE PROGRAM BASICS. Contents

SEMIDEFINITE PROGRAM BASICS. Contents SEMIDEFINITE PROGRAM BASICS BRIAN AXELROD Abstract. A introduction to the basics of Semidefinite programs. Contents 1. Definitions and Preliminaries 1 1.1. Linear Algebra 1 1.2. Convex Analysis (on R n

More information

On the sufficiency of finite support duals in semi-infinite linear programming

On the sufficiency of finite support duals in semi-infinite linear programming On the sufficiency of finite support duals in semi-infinite linear programming Amitabh Basu a, Kipp Martin b, Christopher Thomas Ryan b a The Johns Hopkins University b University of Chicago, Booth School

More information

ON THE UNIQUENESS PROPERTY FOR PRODUCTS OF SYMMETRIC INVARIANT PROBABILITY MEASURES

ON THE UNIQUENESS PROPERTY FOR PRODUCTS OF SYMMETRIC INVARIANT PROBABILITY MEASURES Georgian Mathematical Journal Volume 9 (2002), Number 1, 75 82 ON THE UNIQUENESS PROPERTY FOR PRODUCTS OF SYMMETRIC INVARIANT PROBABILITY MEASURES A. KHARAZISHVILI Abstract. Two symmetric invariant probability

More information

5. Duality. Lagrangian

5. Duality. Lagrangian 5. Duality Convex Optimization Boyd & Vandenberghe Lagrange dual problem weak and strong duality geometric interpretation optimality conditions perturbation and sensitivity analysis examples generalized

More information

FUNCTIONAL ANALYSIS LECTURE NOTES: COMPACT SETS AND FINITE-DIMENSIONAL SPACES. 1. Compact Sets

FUNCTIONAL ANALYSIS LECTURE NOTES: COMPACT SETS AND FINITE-DIMENSIONAL SPACES. 1. Compact Sets FUNCTIONAL ANALYSIS LECTURE NOTES: COMPACT SETS AND FINITE-DIMENSIONAL SPACES CHRISTOPHER HEIL 1. Compact Sets Definition 1.1 (Compact and Totally Bounded Sets). Let X be a metric space, and let E X be

More information

5 Set Operations, Functions, and Counting

5 Set Operations, Functions, and Counting 5 Set Operations, Functions, and Counting Let N denote the positive integers, N 0 := N {0} be the non-negative integers and Z = N 0 ( N) the positive and negative integers including 0, Q the rational numbers,

More information

Victoria Martín-Márquez

Victoria Martín-Márquez A NEW APPROACH FOR THE CONVEX FEASIBILITY PROBLEM VIA MONOTROPIC PROGRAMMING Victoria Martín-Márquez Dep. of Mathematical Analysis University of Seville Spain XIII Encuentro Red de Análisis Funcional y

More information

Decomposability and time consistency of risk averse multistage programs

Decomposability and time consistency of risk averse multistage programs Decomposability and time consistency of risk averse multistage programs arxiv:1806.01497v1 [math.oc] 5 Jun 2018 A. Shapiro School of Industrial and Systems Engineering Georgia Institute of Technology Atlanta,

More information

Iowa State University. Instructor: Alex Roitershtein Summer Homework #1. Solutions

Iowa State University. Instructor: Alex Roitershtein Summer Homework #1. Solutions Math 501 Iowa State University Introduction to Real Analysis Department of Mathematics Instructor: Alex Roitershtein Summer 015 EXERCISES FROM CHAPTER 1 Homework #1 Solutions The following version of the

More information

Lecture 7 Monotonicity. September 21, 2008

Lecture 7 Monotonicity. September 21, 2008 Lecture 7 Monotonicity September 21, 2008 Outline Introduce several monotonicity properties of vector functions Are satisfied immediately by gradient maps of convex functions In a sense, role of monotonicity

More information

Examples of Dual Spaces from Measure Theory

Examples of Dual Spaces from Measure Theory Chapter 9 Examples of Dual Spaces from Measure Theory We have seen that L (, A, µ) is a Banach space for any measure space (, A, µ). We will extend that concept in the following section to identify an

More information

4. Convex Sets and (Quasi-)Concave Functions

4. Convex Sets and (Quasi-)Concave Functions 4. Convex Sets and (Quasi-)Concave Functions Daisuke Oyama Mathematics II April 17, 2017 Convex Sets Definition 4.1 A R N is convex if (1 α)x + αx A whenever x, x A and α [0, 1]. A R N is strictly convex

More information

Assignment 1: From the Definition of Convexity to Helley Theorem

Assignment 1: From the Definition of Convexity to Helley Theorem Assignment 1: From the Definition of Convexity to Helley Theorem Exercise 1 Mark in the following list the sets which are convex: 1. {x R 2 : x 1 + i 2 x 2 1, i = 1,..., 10} 2. {x R 2 : x 2 1 + 2ix 1x

More information

Subdifferential representation of convex functions: refinements and applications

Subdifferential representation of convex functions: refinements and applications Subdifferential representation of convex functions: refinements and applications Joël Benoist & Aris Daniilidis Abstract Every lower semicontinuous convex function can be represented through its subdifferential

More information

Introduction to Convex and Quasiconvex Analysis

Introduction to Convex and Quasiconvex Analysis Introduction to Convex and Quasiconvex Analysis J.B.G.Frenk Econometric Institute, Erasmus University, Rotterdam G.Kassay Faculty of Mathematics, Babes Bolyai University, Cluj August 27, 2001 Abstract

More information

Sum of two maximal monotone operators in a general Banach space is maximal

Sum of two maximal monotone operators in a general Banach space is maximal arxiv:1505.04879v1 [math.fa] 19 May 2015 Sum of two maximal monotone operators in a general Banach space is maximal S R Pattanaik, D K Pradhan and S Pradhan May 20, 2015 Abstract In a real Banach space,

More information

Existence of Global Minima for Constrained Optimization 1

Existence of Global Minima for Constrained Optimization 1 Existence of Global Minima for Constrained Optimization 1 A. E. Ozdaglar 2 and P. Tseng 3 Communicated by A. Miele 1 We thank Professor Dimitri Bertsekas for his comments and support in the writing of

More information

1 Review of last lecture and introduction

1 Review of last lecture and introduction Semidefinite Programming Lecture 10 OR 637 Spring 2008 April 16, 2008 (Wednesday) Instructor: Michael Jeremy Todd Scribe: Yogeshwer (Yogi) Sharma 1 Review of last lecture and introduction Let us first

More information