A Representation of Generalized Convex Polyhedra and Applications
Nguyen Ngoc Luan, Nguyen Dong Yen

TL;DR
This paper extends the representation formulas for convex polyhedra to generalized convex polyhedra in locally convex spaces, enabling new solution existence results for generalized linear programming and vector optimization problems.
Contribution
It develops new representation formulas for generalized convex polyhedra in locally convex spaces, generalizing prior Banach space results.
Findings
Representation formulas for generalized convex polyhedra in locally convex spaces.
Application of formulas to prove solution existence in generalized linear programming.
Application to generalized linear vector optimization problems.
Abstract
It is well known that finite-dimensional polyhedral convex sets can be generated by finitely many points and finitely many directions. Representation formulas in this spirit are obtained for convex polyhedra and generalized convex polyhedra in locally convex Hausdorff topological vector spaces. Our results develop those of X. Y. Zheng (Set-Valued Anal., Vol. 17, 2009, 389-408), which were established in a Banach space setting. Applications of the representation formulas to proving solution existence theorems for generalized linear programming problems and generalized linear vector optimization problems are shown.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
A Representation of Generalized Convex Polyhedra and Applications111This work was supported by National Foundation for Science Technology Development (Vietnam) under the grant No. 101.01-2014.37. The second author thanks the Vietnam Institute for Advanced Study in Mathematics for supporting his 6-month stay at the Institute in 2015.
Nguyen Ngoc Luan222Department of Mathematics and Informatics, Hanoi National University of Education, 136 Xuan Thuy, Hanoi, Vietnam; email: [email protected]. and Nguyen Dong Yen333Institute of Mathematics, Vietnam Academy of Science and Technology, 18 Hoang Quoc Viet, Hanoi 10307, Vietnam; email: [email protected].
Abstract. It is well known that finite-dimensional polyhedral convex sets can be generated by finitely many points and finitely many directions. Representation formulas in this spirit are obtained for convex polyhedra and generalized convex polyhedra in locally convex Hausdorff topological vector spaces. Our results develop those of X. Y. Zheng (Set-Valued Anal., Vol. 17, 2009, 389–408), which were established in a Banach space setting. Applications of the representation formulas to proving solution existence theorems for generalized linear programming problems and generalized linear vector optimization problems are shown.
Mathematics Subject Classification (2010). 49N10, 90C05, 90C29, 90C48.
Key Words. Convex polyhedron, generalized convex polyhedron, locally convex Hausdorff topological vector space, representation formula, generalized linear programming problem, generalized linear vector optimization problem, solution existence theorem.
1 Introduction
The intersection of a finite number of closed half-spaces of a finite-dimensional Euclidean space is called a polyhedral convex set (a convex polyhedron in brief). By convention, the intersection of an empty family of closed half-spaces is the whole space. Therefore, emptyset and the whole spaces are two special polyhedra. Due to [11, Theorem 19.1], for every given convex polyhedron one can find a finite number of points and a finite number of directions such that the polyhedron can be represented as the sum of the convex hull of those points and the convex cone generated by those directions. The converse is also true. This celebrated result is attributed [11, p. 427] primarily to Minkowski [10] and Weyl [13, 14]. By using the result, it is easy to derive fundamental solution existence theorems in linear progamming. Note that the just cited representation formula for finite-dimensional polyhedral convex sets has many other applications in mathematics. As an example, one can refer to the elegant proofs of the necessary and sufficient second-oder conditions for a local solution and for a locally unique solution in quadratic programming, which were given by Contesse [2] in 1980; see [7, pp. 50–63] for details.
According to Bonnans and Shapiro [1, Definition 2.195], a subset of locally convex Hausdorff topological vector space is called a generalized polyhedral convex set (or a generalized convex polyhedron) if it is the intersection of finitely many closed half-spaces and a closed affine subspace of that topological vector space. If the affine subspace can be chosen as the whole space, the generalized polyhedral convex set is said to be a polyhedral convex set (or a convex polyhedron). The theories of generalized linear programming and quadratic programming in [1, Sections 2.5.7 and 3.4.3] are based on this concept of generalized convex polyhedron.
In 2009, using a result related to the Banach open mapping theorem (see, e.g., [12, Theorem 5.20]), Zheng [15, Corollary 2.1] has clarified the relationships between convex polyhedra in Banach spaces and the finite-dimensional convex polyhedra.
It is well known that any infinite-dimensional normed space equipped with the weak topology is not metrizable, but it is a locally convex Hausdorff topological vector space. Similarly, the dual space of any infinite-dimensional normed space equipped with the weak∗ topology is not metrizable, but it is a locally convex Hausdorff topological vector space. Actually, the just mentioned two models provide us with the most typical examples of locally convex Hausdorff topological vector spaces, whose topologies cannot be given by norms. It is clear that Zheng’s results in [15] cannot be used neither for a infinite-dimensional normed space equipped with the weak topology, nor for the dual space of any infinite-dimensional normed space equipped with the weak∗ topology.
The aim of our paper is twofold: to find an analogue of the above-mentioned representation of finite-dimensional convex polyhedra via finite families of points and directions for convex polyhedra in locally convex Hausdorff topological vector spaces, and to apply the obtained results to proving solution existence theorems for infinite-dimensional linear programming problems and linear vector optimization problems. Among other things, we will show that the result of Zheng [15, Corollary 2.1] is valid for convex polyhedra in locally convex Hausdorff topological vector spaces.
The organization of the present paper is as follows. In Section 2, we obtain representation formulas for generalized convex polyhedra. Section 3 is devoted to solution existence of generalized linear programs. Solution existence of generalized linear vector optimization problems is studied in Section 4.
2 Representation Formulas for Generalized Convex Polyhedra
Let be a locally convex Hausdorff topological vector space with the dual space denoted by . For any and , indicates the value of at .
Definition 2.1
(See [1, p. 133]) A subset is said to be a generalized polyhedral convex set (a generalized convex polyhedron for short) if there exist , , , and a closed affine subspace , such that
[TABLE]
If admits the last representation for and for some , , , then it is called a polyhedral convex set (or a convex polyhedron).**
The following classical result shows that, for any convex polyhedron in , one can find a finite number of points and a finite number of directions such that the polyhedron can be represented as the sum of the convex hull of those points and the convex cone generated by those directions. The converse is also true.
Theorem 2.2
([11, Theorem 19.1])* For any nonempty set , the following two properties are equivalent:
(a) is a convex polyhedron;
(b) is finitely generated, i.e., can be represented as*
[TABLE]
where , , and , .
From (2.1) it follows that for . A natural question arises: Is there any analogue of the representation (2.1) for convex polyhedra in locally convex Hausdorff topological vector spaces, or not? In order to give an answer in the affirmative to this question, we will need several results from functional analysis. In what follows, is a locally convex Hausdorff topological vector space.
Lemma 2.3
(Closedness of the sum two linear subspaces; see [12, Theorem 1.42])* Suppose and are linear subspaces of , is closed, and has finite dimension. Then is closed.*
Lemma 2.4
(The Hahn-Banach extension theorem; see [12, Theorem 3.6])* If is a continuous linear functional on a linear subspace of , then there exists such that for all .*
The forthcoming lemma follows from a theorem in [12]. A proof is provided here for the sake of clarity of our presentation.
Lemma 2.5
If and are Hausdorff finite-dimensional topological vector spaces of dimension and if is a linear bijective mapping, then is a homeomorphism.
Proof. Let be a basis of the Euclidean space , which is equipped with the natural topology. Let be a basis of . Setting for , we see that is a basis of . Clearly, there is an unique linear bijection satisfying the conditions for all . Similarly, there is an unique linear bijection with for all . By [12, Theorem 1.21(a)], and are homeomorphisms. (Note that the quoted result was obtained for and topological vector spaces over the complex field . Nevertheless, the method of proof is valid for the case of and topological vector spaces over .) Since and by our construction, it follows that both and are continuous mappings.
We are now in a position to extend Corollary 2.1 from the paper of Zheng [15], which was given in a normed spaces setting, to the case of convex polyhedra in locally convex Hausdorff topological vector spaces.
Proposition 2.6
A nonempty subset is a convex polyhedron if only if there exist closed linear subspaces , of and a convex polyhedron such that
[TABLE]
and
[TABLE]
Proof. Necessity: If is a convex polyhedron, then there exist , , , such that
[TABLE]
Let
[TABLE]
Because is a closed linear subspace of finite codimension, one can find a finite-dimensional linear subspace of , such that and . By [12, Theorem 1.21(b)], is closed. Clearly,
[TABLE]
is a convex polyhedron in . It is easy to verify that . The reverse inclusion is also true. Indeed, for each there exist and satisfying . Since
[TABLE]
for all , it follows that ; hence . We have thus proved that .
Sufficiency: Let , be closed subspaces of satisfying the conditions in (2.2). Let be a convex polyhedron in and let be defined by (2.3). Select and , , such that
[TABLE]
Let , for all , be the canonical projection from on the quotient space . It is clear that the operator , for all , is a linear bijective mapping. On one hand, by [12, Theorem 1.41(a)], is a linear continuous mapping. On the other hand, is a homeomorphism by Lemma 2.5. So, the operator is linear and continuous. Put , . Take any with and . It clear that
[TABLE]
for all . Conversely, take any satisfying for all . Let and be such that . Since
[TABLE]
for all , we see that . Hence . It follows that . Therefore is a convex polyhedron in .
The main result of this section is formulated as follows.
Theorem 2.7
A nonempty subset is a generalized convex polyhedron if and only if there exist , , and a closed linear subspace such that
[TABLE]
Proof. Necessity: Suppose that is a generalized convex polyhedron. Then we have
[TABLE]
where is a closed affine subspace, and for . Select a locally convex Hausdorff topological vector space , a continuous linear mapping , and a point such that . Fix an element and set . It is easy to verify that
[TABLE]
As is a convex polyhedron in , by Proposition 2.6 we can find closed linear subspaces and of and a convex polyhedron such that
[TABLE]
and
[TABLE]
Because is closed and is a closed linear subspace of , is a closed linear subspace of . Since is a convex polyhedron of the finite-dimensional space , invoking Theorem 2.2 we can represent as
[TABLE]
where for and for . Therefore
[TABLE]
We have thus found a representation of the form (2.4) for .
Sufficiency: Suppose that is of the form (2.4). Let
[TABLE]
be the linear subspace generated by the vectors . Put
[TABLE]
By Lemma 2.3, is a closed linear subspace of . Because is a closed subspace of finite codimension of , one can find a finite-dimensional linear subspace , such that and . Consider the continuous linear mapping be defined by , where with . We have
[TABLE]
By Lemma 2.2, is a convex polyhedron of . We have . Indeed, if where and , then . So belongs to . Conversely, for any with and , we have
[TABLE]
Since , is a convex polyhedron in by Proposition 2.6. Hence there exist and such that
[TABLE]
According to Lemma 2.4, there exist , , such that for all Therefore
[TABLE]
It follows that is a generalized polyhedral convex set in .
The next example is an illustration for Theorem 2.7.
Example 2.8
Let be the linear space of continuous real valued functions on the interval with the norm defined by . By the Riesz representation theorem (see e.g. [6, Theorem 6, p. 374] and [9, Theorem 1, p. 113]), the dual space of is the normalized space of functions of bounded variation, that is functions of bounded variation, , and is continuous from the left at every point of . Let be defined by
[TABLE]
where in are chosen such that the vectors are linearly independent. The integrals in (2.5) are Riemannian. They equal respectively to the Riemann-Stieltjes integrals (see [6, p. 367]) and , which are given by the -smooth functions , . Set
[TABLE]
for . It is clear that . The Cauchy-Schwarz inequality
[TABLE]
which is valid for any functions , implies that . As the vectors are linearly independent, we must have . Given any , we want to find a representation of form (2.4) for the convex polyhedron
[TABLE]
Let For with , we have
[TABLE]
and
[TABLE]
Since , there exists an unique pair of real numbers satisfying
[TABLE]
Let the point and the directions be defined by
[TABLE]
It is easy to verify that for , and
[TABLE]
Let us show that
[TABLE]
Take any with and . Because
[TABLE]
and
[TABLE]
we have . Now, take any . Put , , and . Note that , and . Since for , we see that . The formula (2.7) has been proved.**
Based on the preceding example, we can easily construct an illustrative example for polyhedral convex sets in locally convex Hausdorff topological vector spaces.
Example 2.9
Keeping all the notations of Example 2.8, we consider with the weak topology. Then is a locally convex Hausdorff topological vector space whose topology is not a norm topology. The analysis given above shows that the set in (2.6) admits the representation (2.4). **
From Theorem 2.7 we can obtain a representation formula for generalized polyhedral convex cones.
Theorem 2.10
A nonempty set is a generalized polyhedral convex cone if and only if there exist , and a closed linear subspace such that
[TABLE]
Proof. Necessity: If is a generalized polyhedral convex cone, then by Theorem 2.7 we can find , , , , and a closed linear subspace such that
[TABLE]
To show that for , it suffices to observe by (2.9) that belongs to for all , because is a cone. Letting , by the closedness of , we get . Since for , and since for all , and , by choosing for by (2.9) we see that admits the representation (2.8) where is replaced by .
Sufficiency: If has the form (2.8) then it is a cone. In addition, is a generalized polyhedral convex set by Theorem 2.7.
3 Solution Existence in Linear Optimization
Consider a generalized linear programming problem
[TABLE]
where, as before, is a locally convex Hausdorff topological vector space, is a generalized polyhedral convex set, . By definition, the recession cone of a convex set is given by
[TABLE]
If is represented in the form (2.4), then .
The following two existence theorems for (LP) are known results. Actually, in combination, they express the contents of Theorem 2.199 from [1]. The latter, in its turn, is a special case of Theorem 2.198 from [1]. The simple proofs given below show how Theorem 2.7 can be used to study the solution existence of generalized linear programs.
Theorem 3.1
(The Eaves-type Existence Theorem; see [1, Theorem 2.199])* If is nonempty, then (LP) has a solution if and only if for every .*
Proof. If (LP) has a solution , then for each it holds that
[TABLE]
because for every . Hence .
Conversely, suppose that for every . Let us represent in the form (2.4). Select an element such that
[TABLE]
By (2.4), for every there exist , , , and such that . Then we have
[TABLE]
Since can be chosen arbitrarily, must be a solution of (LP).
Remark 3.2
If for every , then one says that the functional is copositive on the recession . We called Theorem 3.1 the Eaves-type Existence Theorem in linear optimization to trace back Eaves’ idea [3, Theorem 3 and Corollary 4, p.702] (see also [7, Theorem 2.2]) in using recession cones for existence theorems in quadratic programming.
Theorem 3.3
(The Frank–Wolfe-type Existence Theorem; see [1, Theorem 2.199])* If is nonempty, then (LP) has a solution if and only if there is a real number such that for every .*
Proof. The necessity is obvious. To prove the sufficiency, suppose that there is a such that for all . Then, for any and we have
[TABLE]
for every . It follows that for any . So, by Theorem 3.1, we can assert that (LP) has a solution.
Remark 3.4
Due to the formulation of the existence theorem in quadratic programming of Frank and Wolfe [4, p. 158] (see also [7, Theorem 2.1]), we called Theorem 3.3 the Frank–Wolfe-type Existence Theorem in linear optimization.
We are interested in studying the region of all for which (LP) has a nonempty solution set, assuming that the constraint set is nonempty and fixed.
Proposition 3.5
If has the form (2.4), then is a generalized polyhedral convex cone of which has the representation
[TABLE]
where is the annihilator [9, p. 117] of .
Proof. By Theorem 3.1, . Therefore, given any , we have for all . Hence, for every one has because . Thus . In addition, for each , one has as . This establishes the inclusion “” in (3.11).
Conversely, suppose that . Since , the last inclusion implies that , for all . Hence, by Theorem 3.1 we can conclude that . The inclusion “” in (3.11) has been proved.
From (3.11) it follows that is a generalized polyhedral convex set. The fact that is a cone is obvious.
Next, for each , we want to describe the solution set of (LP), which is denoted by . For doing so, let us suppose that is given by (2.4) and consider the index sets
[TABLE]
and
[TABLE]
Note that is nonempty, but it may happen that is empty.
Proposition 3.6
If and is given by (2.4), then
[TABLE]
In particular, is a generalized polyhedral convex set.
Proof. First, take an arbitrary element from the set on the right-hand-side of (3.12). Let
[TABLE]
where for all , , for all , and . By (2.4), for each one can find for , , for , and such that
[TABLE]
By Proposition 3.5, . If , then using Theorem 3.1 and formula we get
[TABLE]
Now, selecting an index and recalling the definition , we get
[TABLE]
It follows that . We have shown that .
Second, take any vector and represent it in the form
[TABLE]
where for , , for , and . It is easy to show that for all and for all . This implies that belongs to the set on the right-hand-side of (3.12).
The proof is complete.
4 The Weakly Efficient Solution Set in Linear Vector Optimization
Consider a generalized linear vector optimization problem of the form
[TABLE]
with being a continuous linear mapping between locally convex Hausdorff topological vector spaces, a generalized polyhedron, a polyhedral convex cone.
We say that is a weakly efficient solution of (VLP) if there does not exist any such that . The set of all the weakly efficient solutions is denoted by . We are interested in finding conditions to have .
By a standard scalarization scheme in vector optimization, given any , we define the scalar problem
[TABLE]
To make our presentation easier for reading, we give simple proof for the following known result.
Lemma 4.1
(See [8, Proposition 3.2, p. 95])* If is nonempty, then is a weakly efficient solution of (VLP) if and only if there exists , where , such that*
[TABLE]
Proof. First, suppose that . Since \big{(}Mu-M(D)\big{)}\cap{\rm int}K=~{}\emptyset and since and are convex sets, by the separation theorem (see, e.g., [1, Theorem 2.13]), there exists such that
[TABLE]
for all and . Substituting to the above inequality yields for all . Hence . Choosing , one has for every . This shows that the inclusion (4.13) is valid.
Now, suppose that and there is such that (4.13) is satisfied. If , then there exist and a balanced neighborhood of 0 satisfying . Hence, for each , one has . In combination with the inequality which is guaranteed by (4.13), this implies . As is a balanced neighborhood of 0, we can assert that for all . Let be such that . Since there exists with , we get , a contradiction.
Remark 4.2
Looking back to the proof of Lemma 4.1, we can observe that it suffices to assume that is a convex cone. In other words, the polyhedrality of is superfluous for the assertion of the lemma.**
We have for all , where is the adjoint operator of .
Theorem 4.3
Problem (VLP) has a solution if and only if
[TABLE]
In particular, if
[TABLE]
then (VLP) has a solution.
Proof. By Lemma 4.1, (VLP) has a solution if and only if there exists such that the solution set of is non empty. According to Theorem 3.1, this solution set is non-void if and only if . Thus, we have shown that (VLP) has a solution if and only if (4.14) is fulfilled.
Now, suppose that (4.15) is satisfied. Then we can find such that and . Since the later obviously implies that , we have . Hence (4.14) holds true, so (VLP) has a solution.
Remark 4.4
The assertions of Theorem 4.3 are valid for the case is an arbitrary convex cone.**
We conclude this section by a statement about the structure of which is applicable also the case is an arbitrary convex cone.
Theorem 4.5
The weakly efficient solution set of (VLP) is the union of finitely many generalized polyhedral convex sets.
Proof. Using Lemma 4.1, we can represent the weakly efficient solution set of (VLP) as follows
[TABLE]
Setting , we can rewrite (4.16) as
[TABLE]
where is the solution set of the problem (LP) considered in Section 3. Invoking (3.12) and noting that the number of the index sets (resp., the number of the index sets ) is finite, from (4.17) we obtain the desired conclusion.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] J. F. Bonnans and A. Shapiro, Perturbation Analysis of Optimization Problems , Springer, New York, 2000.
- 2[2] L. Contesse, Une caractérisation complète des minima locaux en programmation quadratique , Numerische Mathematik 34 (1980), 315–332.
- 3[3] B. C. Eaves, On quadratic programming , Management Science 17 (1971), 698–711.
- 4[4] M. Frank and P. Wolfe, An algorithm for quadratic programming , Naval Research Logistics Quarterly 3 (1956), 95–110.
- 5[5] A. D. Ioffe and V. M. Tihomirov, Theory of Extremal Problems , North-Holland Publishing Company, Amsterdam, 1979.
- 6[6] A. N. Kolmogorov and S. V. Fomin, Introductory Real Analysis , Dover Publications, New York, 1975.
- 7[7] G. M. Lee, N. N. Tam, and N. D. Yen, Quadratic Programming and Affine Variational Inequalities: A Qualitative Study , Springer-Verlag, New York, 2005.
- 8[8] D. T. Luc, Theory of Vector Optimization , Springer-Verlag, Berlin-Heidelberg, 1989.
