Optimality Conditions For Set-Valued Optimization Problems Homework

Abstract

The vector criterion and set criterion are two defining approaches of solutions for the set-valued optimization problems. In this paper, the optimality conditions of both criteria of solutions are established for the set-valued optimization problems. By using Studniarski derivatives, the necessary and sufficient optimality conditions are derived in the sense of vector and set optimization.

Keywords: vector optimization, set optimization, optimality conditions, pseudoconvexity, Studniarski derivative

Introduction

Let F be a set-valued mapping from a set S to a real normed linear space Y and let D ⊂ Y be a convex cone. The set-valued optimization problem is formalized as follows:

It is generally known that there are two types of criteria of solutions for problem (P): the vector criteria and set criteria. The vector criterion is the most commonly used in problem (P), which is looking for efficient elements of the set F(S) = ⋃xSF(x). The problem (P) with vector criterion is often defined as set-valued vector optimization. Over the past decades, the set-valued vector optimization theory and its applications have been investigated extensively by many authors. Fruitful results are produced in this field, and we recommend the excellent books [1–4] and the references therein to the reader.

Another criterion, the set criterion, also called set optimization, was proposed by Kuroiwa [5, 6] in 1999 for the first time. Set optimization consists of looking for image sets , with , which satisfy some set-relations between the rest of the image sets F(x) with x ∈ S. In [3], p. 378, Jahn states that the set relation approach opens a new and exciting field of research. Although the set criterion seems to be more natural and interesting than the traditional one, the study of set optimization is very limited, such as papers on existence conditions, see [7–12]; on duality theory, see [13–16]; on optimality conditions, see [17–19]; on scalarization [20–23]; on well-posedness properties, see [24, 25]; on Ekeland variational principles, see [26, 27].

The investigation of optimality conditions, especially as regards the vector criterion, has received tremendous attention in the research of optimization problems and has been studied extensively. As everyone knows most of the problems under consideration are nonsmooth, which leads to introducing many kinds of generalized derivatives by the authors. A meaningful concept is the Studniarski derivative [28], which has some properties similar to the contingent derivative [1] and the Hadamard derivative [29]. Recently, much attention has been paid to optimality conditions and related topics for vector optimization by using Studniarski derivatives; see [30, 31].

Inspired by the above observations, in this paper, the optimality conditions are established for problem (P) in both the vector criterion and set criterion by using Studniarski derivatives. The rest of the paper is organized as follows: In Section 2, some well-known definitions and results used in sequel are recalled. In Section 3 and Section 4, the necessary and sufficient optimality conditions are given in the sense of the vector criterion and set criterion, respectively.

Preliminaries

Throughout the rest of paper, it is assumed that X and Y are two real normed linear spaces, and C ⊂ Y is a convex closed pointed cones with nonempty interior, i.e., intC ≠ ∅. The partial order of Y is deduced by cone C. The ordering given by C on Y is denoted by ⩽, which is defined as follows:

y1 ⩽ y2 if y2 − y1 ∈ C.

Let S ⊂ X be a nonempty set and . The contingent cone to S at is defined by (see [1]):

For a set-valued mapping F:X → 2Y, the set

dom(F): = {x ∈ X:F(x) ≠ ∅}, 

is called the domain of F, the set

graph(F): = {(xy) ∈ X × Y: y ∈ F(x)}

is called the graph of the map F, and the set

epi(F): = {(xy) ∈ X × Y: y ∈ F(x) + C}

is termed the epigraph of F. Let , based upon the definition of contingent cone, we can derive , which is formulated by

2.1

Definition 2.1

see [1]

Let F:X → 2Y be a set-valued map. F is called C-pseudoconvex at if

For a set-valued mapping F:X → 2Y, the upper and lower Studniarski derivatives of the map F is defined as follows:

Definition 2.2

see [28]

Let .

  • (i)

    The upper Studniarski derivatives of F at is defined by
  • (ii)

    The lower Studniarski derivative of F at is defined by

Remark 2.3

Regarding the upper and lower Studniarski derivatives of the map F, one has:

  1. It is easily to see from Definition 2.2 that
  2. The upper Studniarski derivative is an exactly contingent derivative [1] or an upper Hadamard derivative [29], and the lower Studniarski derivative is the lower Hadamard derivative introduced by Penot [29].

Lemma 2.4 is an useful property associated with C-pseudoconvexity of a set-valued mapping, which will be used in the next two sections.

Lemma 2.4

LetS ⊂ Xbe a convex set andF:S → 2YbeC-pseudoconvex at. One has

2.2

Proof

Since F at is C-pseudoconvex, it yields

Therefore, we see from (2.1) that there exist (tn) → 0+, such that

Thus

Then one has

By taking the limits when n →  + ∞, we derive

So, we get (2.2), as desired. □

Example 2.5

Let XY = ℝ, C = ℝ+ and F:X → 2Y be defined by

F(x): = {y ∈ Y:y ≥ x2},  for all x ∈ X.

Taking , we can derive that

On the other hand, it is obviously that F is C-pseudoconvex at and satisfies

Optimality conditions with vector criterion

Let S ⊂ X be a nonempty set, F:S → 2Y be a set-valued mapping and .

From now on, for convention we always assume that the upper and lower Studniarski derivatives of the mapFatexist and.

Consider the set-valued optimization problem (P) formulated in Section 1. We shall establish the optimality conditions for vector criteria of solutions for problem (P). In order to distinguish from the case of set criteria, we rewrite the problem (P) as follows:

Definition 3.1

The pair is said to be a weak minimizer of (VP), denoted by , if

Theorems 3.2 and 3.3 are necessary optimality conditions for the weak minimizer of problem (VP).

Theorem 3.2

Letbe a weak minimizer of (VP). Then

3.1

Proof

Let with x ∈ S. Then there exist (tn) → 0+, and such that

3.2

Because is a weak minimizer of (VP), one has

3.3

Now, let us show that y ∉ −intC. In fact, assuming that y ∈ −intC, since −intC is open, we obtain from (3.2)

Hence, it follows from that

which contradicts (3.3). □

Theorem 3.3

Let. Ifis a weak minimizer of (VP), then

3.4

Proof

According to Theorem 3.2, one shows that for all x ∈ S

3.5

We proceed by contradiction. If the conclusion in Theorem 3.3 does not hold, then there exists such that

3.6

Noticing that , there exist (tn) → 0+ and (xn) → x such that for all n. In addition, we see from (3.6) that there exists . Hence, with the above (tn) and (xn) there is such that

So, we see that there exist (tn) → 0+, (xn) → x, and such that

Thus,

which is a contradiction to (3.5). □

Now, we present a sufficient optimality condition for problem (VP) under the assumption of pesudoconvexity of F.

Theorem 3.4

LetSbe a convex set andFbeC-pseudoconvex at. If

3.7

thenis a weak minimizer of (VP).

Proof

We proceed by contradiction. Suppose that is not a weak minimizer of (VP), then there exists x ∈ S such that

3.8

Since F is C-pseudoconvex at , one shows from Lemma 2.4 that

Therefore, one shows from (3.8) that

and as a consequence

This is a contradiction to (3.7). □

Optimality conditions with set criterion

This section works with the optimality conditions for problem (P) in the sense of the set criterion. Firstly, let us recall the concepts of set relations.

Definition 4.1

see [5]

Let A and B be two nonempty sets of Y. We write AlB, if for all b ∈ B there exists a ∈ A such that a ⩽ b. Here, ‘⩽l’ is called a lower relation.

Remark 4.2

The above defined lower relation is equivalent to B ⊂ AC and it is the generalization of the order induced by a pointed convex cone C in Y, in the following sense: a ⩽ b if b ∈ aC.

Definition 4.3

see [14]

Let A and B be two nonempty sets of Y. We write A<lwB, if B ⊂ A + intC. Here, ‘<lw’ is named a lower weak relation.

Definition 4.4

see [14]

Let 𝒜 be a family of subsets of Y.

  1. A ∈ 𝒜 is named a lower minimal of 𝒜, if for each B ∈ 𝒜 such that BlA, it satisfies AlB;

  2. A ∈ 𝒜 is called a lower weak minimal of 𝒜, if for each B ∈ 𝒜 such that B<lwA, it satisfies A<lwB.

The relation ‘⩽l’ determines in 𝒜 the equivalence relation:

A ∼ B  ⇔  AlB and BlA

whose classes it is denoted by [A]l. Analogously the relation ‘<lw’ determines in 𝒜 an equivalence relation whose classes are written [A]lw.

In the sense of a set criterion, the problem (P) can be rewritten in the next form:

In this case, a lower weak minimizer of the problem (SP) is a pair such that and is a lower weak minimal of the family of images of F, i.e. the family

ℱ = {F(x):x ∈ S}.

We say that is a lower weak minimal of (SP) instead of ℱ.

Example 4.5

Let S = [0, 1] ⊂ ℝ, Y = ℝ2 and . Let F0:[0, 1] → 22 be a set-valued map and defined by

where [yy] with yy ∈ ℝ2 denotes the line segment, i.e.[yy] = {λy + (1 − λ)y:0 ≤ λ ≤ 1}. Considering the following problem (SP)0:

By computing, we can derive that (1, F(1)) is the lower minimizer of (SP)0 and (xF(x)) with 0 < x ≤ 1 is the lower weak minimizer of (SP)0.

Definition 4.6

see [19]

Let be a lower weak minimal of (SP). is named strict lower weak minimal of (SP), if there exists a neighborhood of such that , for all . Then is called a strict lower weak minimum of (SP).

Lemma 4.7

see [19]

Let. Thenis a lower weak minimal of (SP) if and only if for eachx ∈ Sone of the conditions below is fulfilled:

  1. .

  2. There existssuch that.

Definition 4.8

see [19]

(Domination property). It is called that a subset A ⊂ Y has the D-weak minimal property, if for all y ∈ A there exists a weakly minimal a of A such that a − y ∈ (−intC) ∪ {0}.

Lemma 4.9

Letand. If, andhas theC-weak minimal property, one has

4.1

Proof

Assuming that (4.1) does not hold, that is,

Then there exist y ∈ F(x) and d ∈ intC such that . So, it yields and . On the other hand, it follows from and the C-weak minimal property of that, for all , there is d ∈ intC ∪ {0} such that

Therefore, we derive that

which means . This contradicts the hypothesis. □

Theorem 4.10

Letbe a strict lower weak minimum of (SP). Ifandhas theC-weak minimal property, then

4.2

Proof

Because is a strict lower weak minimum of (SP), we see from Definition 4.6 that there exists a neighborhood of such that for all . Let x ∈ S. Assuming that then there exist (tn) → 0+, and such that

Therefore for large enough n, we can get which verifies . Thus, it follows from Lemma 4.9 that

4.3

Let us prove that y ∉ −intD. Otherwise, one has

Noticing that , we derive

Как бы там ни было, Стратмор знал, что Хейла можно будет всегда ликвидировать в случае необходимости. - Решайтесь, приятель! - с издевкой в голосе сказал Хейл.  - Мы уходим или нет? - Его руки клещами сжимали горло Сьюзан. Стратмор знал, что, если он сейчас достанет мобильник и позвонит в службу безопасности, Сьюзан будет жить. Он готов был спорить на что угодно, хоть на собственную жизнь, потому что ясно представлял себе весь сценарий.

0 comments

Leave a Reply

Your email address will not be published. Required fields are marked *