The aim of this paper is to introduce a new iterative scheme for finding common solutions of the variational inequalities for an inverse strongly accretive mapping and the solutions of fixed point problems for nonexpansive semigroups by using the modified viscosity approximation method associate with Meir-Keeler type mappings and obtain some strong convergence theorem in a Banach spaces under some parameters controlling conditions. Our results extend and improve the recent results of Li and Gu (2010), Wangkeeree and Preechasilp (2012), Yao and Maruster (2011), and many others.
1. Introduction
The theory of variational inequalities and variational inclusions are among the most interesting and important mathematical problems and have been studied intensively in the past years since they have wide applications in the optimization and control, economics, engineering science, physical sciences, and applied sciences. For these reasons, many existence result and iterative algorithms for various variational inclusion have been studied extensively by many authors (see, e.g., [1–8]). The important generalization of variational inequalities has been extensively studied and generalized in different directions to study a wide class of problems arising in mechanics, optimization, nonlinear programming, finance, and applied sciences (see, e.g., [2, 8]).
Let C be a nonempty closed convex subset of a real Banach space E and E* be the dual space of E with norm ∥·∥ and 〈·,·〉 pairing between E and E*. For q>1, the generalized duality mapping Jq:E→2E* is defined by
(1.1)Jq(x)={f∈E*:〈x,f〉=∥x∥q,∥f∥=∥x∥q-1}
for all x∈E. In particular, if q=2, the mapping J2 is called the normalized duality mapping and, usually, written J2=J. Further, we have the following properties of the generalized duality mapping Jq: (i) Jq(x)=∥x∥q-2J2(x) for all x∈E with x≠0; (ii) Jq(tx)=tq-1Jq(x) for all x∈E and t∈[0,∞); (iii) Jq(-x)=-Jq(x) for all x∈E.
Recall that a mapping A:C→C is said to be (i) Lipschitzian with Lipschitz constant L>0 if ∥Ax-Ay∥≤L∥x-y∥, for all x,y∈C; (ii) contraction if there exists a constant α∈(0,1) such that ∥Ax-Ay∥≤α∥x-y∥, for all x,y∈C; (iii) nonexpansive if ∥Ax-Ay∥≤∥x-y∥, for all x,y∈C. An operator A:C→E is said to be
accretive if there exists j(x-y)∈J(x-y) such that
(1.2)〈Ax-Ay,j(x-y)〉≥0,∀x,y∈C,
β-strongly accretive if there exists a constant β>0 such that
(1.3)〈Ax-Ay,j(x-y)〉≥β∥x-y∥2,∀x,y∈C,
β-inverse strongly accretive if, for any β>0,
(1.4)〈Ax-Ay,j(x-y)〉≥β∥Ax-Ay∥2,∀x,y∈C.
Let D be a subset of C and Q:C→D. Then Q is said to sunny if Q(Qx+t(x-Qx))=Qx, whenever Qx+t(x-Qx)∈C for x∈C and t≥0. A subset D of C is said to be a sunny nonexpansive retract of C if there exists a sunny nonexpansive retraction Q of C onto D. A mapping Q:C→C is called a retraction if Q2=Q. If a mapping Q:C→C is a retraction, then Qz=z for all z is in the range of Q.
A family 𝒮={T(t):0≤t<∞} of mappings of C into itself is called a nonexpansive semigroup on C if it satisfies the following conditions:
T(0)x=x for all x∈C;
T(s+t)=T(s)T(t) for all s,t≥0;
∥T(t)x-T(t)y∥≤∥x-y∥ for all x,y∈C and t≥0;
for all x∈C,t↦T(t)x is continuous.
We denote by F(𝒮) the set of all common fixed points of 𝒮={T(t):0≤t<∞},thatis,F(𝒮)=∩t≥0F(T(t)). It is known that F(𝒮) is closed and convex (see also [9, 10]).
A mapping ψ:ℝ+→ℝ+ is said to be an L function if ψ(0)=0,ψ(t)>0 for each t>0, and for every s>0 there exists u>s such that ψ(t)≤s for all t∈[s,u]. As a consequence, every L-function ψ satisfies ψ(t)<t for each t>0.
Definition 1.1.
Let (X,d) be a matric space. A mapping f:X→X is said to be:
(ψ,L)-contraction if ψ:ℝ+→ℝ+ is an L-function and d(f(x),f(y))<ψ(d(x,y)) for all x,y∈X with x≠y;
Meir-Keeler type mapping if for each ϵ>0 there exists δ=δ(ϵ)>0 such that for each x,y∈X with d(x,y)<ϵ+δ we have d(f(x),f(y))<ϵ.
Remark 1.2.
From Definition 1.1, if ψ(t)=αt,α∈[0,1), t∈ℝ+, then we get the usual contraction mapping with coefficient α.
At the same time, we are also interesting in the variational inequality problems for an inverse strongly accretive mappings in Banach spaces. In 2006, Aoyama et al. [11] introduced the following iteration scheme for an inverse strongly accretive operator A in Banach spaces E:
(1.5)x1=x∈C,xn+1=αnxn+(1-αn)QC(xn-λnAxn),
for all n≥1, where C⊂E and QC is a sunny nonexpansive retraction from E onto C. They proved a weak convergence theorem in a Banach spaces. Moreover, the sequence {xn} in (1.5) solved the generalized variational inequality problem for finding a point x∈C such that
(1.6)〈Ax,j(y-x)〉≥0
for all y∈C. The set of solutions of (1.6) is denoted by VI(C,A).
An interesting is the proof by using a nonexpansive semigroup and Meir-Keeler type mapping, in 2010, Li and Gu [12] defined the following sequence:
(1.7)x1=x∈E,yn-αnxn+(1-αn)T(tn)xn,xn+1=βnf(xn)+(1-βn)yn,n≥1.
Wangkeeree and Preechasilp [13] introduced the following iterative scheme:
(1.8)x0∈C,zn-γnxn+(1-γn)T(tn)xn,yn-αnxn+(1-αn)T(tn)zn,xn+1=βnf(xn)+(1-βn)yn,n≥0.
In 2011, Yao and Maruster [8] proved some strong convergence theorems for finding a solution of variational inequality problem (1.6) in Banach spaces. They defined a sequence {xn} iteratively by given arbitrarily x0∈C and
(1.9)xn+1=βnxn+(1-βn)QC[(1-αn)(xn-λAxn)],∀n≥0,
where QC is a sunny nonexpansive retraction from a uniformly convex and 2-uniformly smooth Banach space E, and A is an α-inverse strongly accretive operator of C into E.
Motivated and inspired by the idea of Li and Gu [12], Wangkeeree and Preechasilp [13], and Yao and Maruster [8], in this paper, we introduce a new iterative scheme for finding common solutions of the variational inequalities for an inverse strongly accretive mapping and the solutions of fixed point problems for a nonexpansive semigroup by using the modified viscosity approximation method associated with Meir-Keeler type mapping. We will prove the strong convergence theorem under some parameters controlling conditions. Our results extend and improve the recent results of Li and Gu [12], Wangkeeree and Preechasilp [13], Yao and Maruster [8], and many others.
2. Preliminaries
Let U={x∈E:∥x∥=1}. A Banach space E is said to uniformly convex if, for any ϵ∈(0,2], there exists δ>0 such that, for any x,y∈U, ∥x-y∥≥ϵ implies ∥(x+y)/2∥≤1-δ. It is known that a uniformly convex Banach space is reflexive and strictly convex. A Banach space E is said to be smooth if the limit limt→0((∥x+ty∥-∥x∥)/t) exists for all x,y∈U. It is also said to be uniformly smooth if the limit is attained uniformly for x,y∈U. The modulus of smoothness of E is defined by
(2.1)ρ(τ)=sup{12(∥x+y∥+∥x-y∥)-1:x,y∈E,∥x∥=1,∥y∥=τ},
where ρ:[0,∞)→[0,∞) is a function. It is known that E is uniformly smooth if and only if limτ→0(ρ(τ)/τ)=0. Let q be a fixed real number with 1<q≤2. A Banach space E is said to be q-uniformly smooth if there exists a constant c>0 such that ρ(τ)≤cτq for all τ>0.
A Banach space E is said to satisfy Opial’s condition if for any sequence {xn} in E, xn⇀x(n→∞) implies that
(2.2)limsupn→∞∥xn-x∥<limsupn→∞∥xn-y∥,∀y∈Ewithx≠y.
By [14, Theorem 1], it is well known that if E admits a weakly sequentially continuous duality mapping, then E satisfies Opial’s condition, and E is smooth.
The following result describes a characterization of sunny nonexpansive retractions on a smooth Banach space.
Proposition 2.1 (see [15]).
Let E be a smooth Banach space and let C be a nonempty subset of E. Let Q:E→C be a retraction, and let J be the normalized duality mapping on E. Then the following are equivalent:
Q is sunny and nonexpansive;
∥Qx-Qy∥2≤〈x-y,J(Qx-Qy)〉,∀x,y∈E;
〈x-Qx,J(y-Qx)〉≤0,∀x∈E,y∈C.
Proposition 2.2 (see [16]).
Let C be a nonempty closed convex subset of a uniformly convex and uniformly smooth Banach space E, and let T be a nonexpansive mapping of C into itself with F(T)≠∅. Then the set F(T) is a sunny nonexpansive retract of C.
Lemma 2.3 (see [17]).
Let C be a nonempty bounded closed convex subset of a uniformly convex Banach space E and T:C→C be a nonexpansive mapping. If {xn} is a sequence of C such that xn⇀x and xn-Txn→0, then x is a fixed point of T.
We need the following lemmas for proving our main results.
Lemma 2.4 (see [18]).
Let r>0, and let E be a uniformly convex Banach space. Then, there exists a continuous, strictly increasing, and convex function g:[0,∞)→[0,∞) with g(0)=0 such that
(2.3)∥λx+(1-λ)y∥2≤λ∥x∥2+(1-λ)∥y∥2-λ(1-λ)g(∥x-y∥)
for all x,y∈Br:={z∈E:∥z∥≤r} and 0≤λ≤1.
Lemma 2.5 (see [19]).
Let E be a real smooth and uniformly convex Banach space, and let r>0. Then there exists a strictly increasing, continuous, and convex function g:[0,2r]→ℝ such that g(0)=0 and
(2.4)g(∥x-y∥)≤∥x∥2-2〈x,jy〉+∥y∥2,∀x,y∈Br.
Lemma 2.6 (see [18]).
Let E be a real 2-uniformly smooth Banach space with the best smooth constant K. Then the following inquality holds:
(2.5)∥x+y∥2≤∥x∥2+2〈y,Jx〉+2∥Ky∥2,∀x,y∈E.
Lemma 2.7 (see [20]).
Let E be a real Banach space and J:E→2E* be the normalized duality mapping. Then, for any x,y∈E, one has
(2.6)∥x+y∥2≤∥x∥2+2〈y,J(x+y)〉
for all j(x+y)∈J(x+y) with x≠y.
Lemma 2.8 (see [21]).
Let {xn} and {yn} be bounded sequences in a Banach space X, and let {βn} be a sequence in [0,1] with 0<liminfn→∞βn≤limsupn→∞βn<1. Suppose that xn+1=(1-βn)yn+βnxn for all integers n≥0 and limsupn→∞(∥yn+1-yn∥-∥xn+1-xn∥)≤0. Then, limn→∞∥yn-xn∥=0.
Lemma 2.9 (see [22]).
Assume that {an} is a sequence of nonnegative real numbers such that
(2.7)an+1≤(1-αn)an+δn,n≥0,
where {αn} is a sequence in (0,1), and {δn} is a sequence in ℝ such that
∑n=1∞αn=∞
limsupn→∞(δn/αn)≤0 or ∑n=1∞|δn|<∞.
Then limn→∞an=0.
Theorem 2.10 (see [23]).
Let (X,d) be a complete metric space and f:X→X a Meir-Keeler type mapping. Then f has a unique fixed point.
Theorem 2.11 (see [24]).
Let (X,d) be a metric space and f:X→X a mapping. Then the following assertions are equivalent:
f is a Meir-Keeler type mapping;
there exists an L function ψ:ℝ+→ℝ+ such that f is a (ψ,L) contraction.
Proposition 2.12 (see [21]).
Let C be a convex subset of a Banach space E. Let f:C→C be a Meir-Keeler type mapping. Then for each ϵ>0 there exists r∈(0,1) such that for each x,y∈C with ∥x-y∥≥ϵ, one has
(2.8)∥f(x)-f(y)∥≤r∥x-y∥.
Proposition 2.13 (see [21]).
Let C be a convex subset of a Banach space E. Let T be a nonexpansive mapping on C, and let f:C→C be a Meir-Keeler-type mapping. Then the following holds:
T∘f is a Meir-Keeler type mapping on C;
for each α∈(0,1) the mapping x↦αf(x)+(1-α)T(x) is a Meir-Keeler-type mapping on C.
The following lemma is characterized by the set of solutions of variational inequality by using sunny nonexpansive retractions.
Lemma 2.14 (see [11]).
Let C be a nonempty closed convex subset of a smooth Banach space E. Let QC be a sunny nonexpansive retraction from E onto C, and let A be an accretive operator of C into E. Then, for all λ>0,
(2.9)VI(C,A)=F(Q(I-λA)),
where VI(C,A)={x*∈C:〈Ax*,J(x-x*)〉≥0,∀x∈C}.
3. Strong Convergence Theorems
In this section, we suppose that the function ψ from the definition of the (ψ,L) contraction is continuous and strictly increasing and limt→∞η(t)=∞, where η(t)=t-ψ(t), t∈ℝ+. In consequence, we have that η is a bijection on ℝ+ and the function ψ satisfies the assumption in Remark 1.2.
Suppose that {αn},{βn}⊂(0,1), and {μn},{λn}⊂(0,∞) satisfy the following conditions:
limn→∞αn=0 and ∑n=0∞αn=∞;
limn→∞|λn+1-λn|=0 and 0<a≤λn≤b<β/K2;
0<liminfn→∞βn≤limsupn→∞βn<1;
limn→∞μn=0;
limn→∞supx∈C~∥T(μn+1)x-T(μn)x∥=0, C~ bounded subset of C.
Next, we stat the main result.
Theorem 3.1.
Let E be a uniformly convex and 2-uniformly smooth Banach space with the best smooth constant K and C a nonempty closed convex subset of E. Let QC be a sunny nonexpansive retraction from E onto C and A:C→E be an β-inverse-strongly accretive operator. Let 𝒮={T(t):t≥0} be a nonexpansive semigroup from C into itself and f be a Meir-Keeler contraction of C into itself. Suppose that F:=F(𝒮)∩VI(C,A)≠∅ and the conditions (C1)–(C5). For arbitrary given x1∈C, the sequences {xn} are generated by
(3.1)un=QC(xn-λnAxn),yn=QC[αnf(xn)+(1-αn)un],xn+1=βnxn+(1-βn)T(μn)yn.
Then {xn} converges strongly to x*=QFfx*∈F which also solves the following variational inequality:
(3.2)〈(f-I)x*,J(z-x*)〉≤0,∀z∈F.
Proof.
First we prove that {xn} bounded. Let p∈F, we have
(3.3)∥un-p∥2=∥QC(xn-λnAxn)-QC(p-λnAp)∥2≤∥(xn-λnAxn)-(p-λnAp)∥2=∥(I-λnA)xn-(I-λnA)p∥2=∥(xn-p)-λn(Axn-Ap)∥2≤∥xn-p∥2-2λn〈Axn-Ap,J(xn-p)〉+2K2λn2∥Axn-Ap∥2≤∥xn-p∥2-2λnβ∥Axn-Ap∥2+2K2λn2∥Axn-Ap∥2=∥xn-p∥2+2λn(λnK2-β)∥Ax-Ay∥2≤∥xn-p∥2.
So, we get ∥un-p∥≤∥xn-p∥, for all n≥1. It follows that
(3.4)∥xn+1-p∥=∥βn(xn-p)+(1-βn)(T(μn)yn-p)∥≤βn∥xn-p∥+(1-βn)∥yn-p∥=βn∥xn-p∥+(1-βn)∥QC[αnf(xn)+(1-αn)un]-QCp∥≤βn∥xn-p∥+(1-βn)[αn∥f(xn)-p∥+(1-αn)∥un-p∥]≤βn∥xn-p∥+(1-βn)[αn∥f(xn)-f(p)∥+αn∥f(p)-p∥+(1-αn)∥xn-p∥]≤βn∥xn-p∥+(1-βn)[αnψ(∥xn-p∥)+αn∥f(p)-p∥+(1-αn)∥xn-p∥]=∥xn-p∥+αn(1-βn)[η(∥xn-p∥)]+αn(1-βn)η[η-1(∥f(p)-p∥)]≤max{∥xn-p∥,η-1(∥f(p)-p∥)}.
By induction, we conclude that
(3.5)∥xn-p∥≤max{∥xn-p∥,η-1(∥f(p)-p∥)},∀n≥1.
This implies that {xn} bounded, so are {f(xn)}, {yn}, {un}, {Axn}, and {T(μn)yn}.
Next, we show that limn→∞∥xn+1-xn∥=0, we observe that
(3.6)∥un+1-un∥=∥QC(xn+1-λn+1Axn+1)-QC(xn-λnAxn)∥≤∥(xn+1-λn+1Axn+1)-(xn-λnAxn)∥=∥(xn+1-λn+1Axn+1)-(xn-λn+1Axn)+(λn+1-λn)Axn∥≤∥(I-λn+1A)xn+1-(I-λn+1A)xn∥+|λn+1-λn|∥Axn∥≤∥xn+1-xn∥+|λn+1-λn|∥Axn∥,∥yn+1-yn∥=∥QC[αn+1f(xn+1)+(1-αn+1)un+1]-QC[αnf(xn)+(1-αn)un]∥≤∥[αn+1f(xn+1)+(1-αn+1)un+1]-[αnf(xn)+(1-αn)un]∥≤∥un+1-un∥+αn+1(∥un+1∥+∥f(xn+1)∥)+αn(∥un∥+∥f(xn)∥)≤∥xn+1-xn∥+|λn+1-λn|∥Axn∥+αn+1(∥un+1∥+∥f(xn+1)∥)+αn(∥un∥+∥f(xn)∥).
It follows that
(3.7)∥T(μn+1)yn+1-T(μn)yn∥≤∥T(μn+1)yn+1-T(μn+1)yn∥+∥T(μn+1)yn-T(μn)yn∥≤∥yn+1-yn∥+∥T(μn+1)yn-T(μn)yn∥≤∥xn+1-xn∥+|λn+1-λn|∥Axn∥+αn+1(∥un+1∥+∥f(xn+1)∥)+αn(∥un∥+∥f(xn)∥)+∥T(μn+1)yn-T(μn)yn∥≤∥xn+1-xn∥+|λn+1-λn|∥Axn∥+αn+1(∥un+1∥+∥f(xn+1)∥)+αn(∥un∥+∥f(xn)∥)+supy∈{yn}∥T(μn+1)y-T(μn)y∥.By (C1), (C2), and (C4), they imply that
(3.8)limsupn→∞(∥T(μn+1)yn+1-T(μn)yn∥-∥xn+1-xn∥)≤0.
Applying Lemma 2.8, we obtain
(3.9)limn→∞∥T(μn)yn-xn∥=0.
Therefore, we have
(3.10)limn→∞∥xn+1-xn∥=0.
On the other hand, we consider
(3.11)∥xn+1-p∥2≤βn∥xn-p∥2+(1-βn)∥yn-p∥2≤βn∥xn-p∥2+(1-βn)[αn∥f(xn)-p∥2+(1-αn)∥un-p∥2]=βn∥xn-p∥2+(1-βn)×[αn∥f(xn)-p∥2+(1-αn)∥QC(xn-λnAxn)-QC(p-λnAp)∥2]≤βn∥xn-p∥2+(1-βn)×[αn∥f(xn)-p∥2+(1-αn)(∥xn-p∥2+2λn(λnK2-β)∥Ax-Ay∥2)]=[βn+(1-βn)(1-αn)]∥xn-p∥2+αn(1-βn)∥f(xn)-p∥2+2λn(λnK2-β)(1-βn)(1-αn)∥Ax-Ay∥2≤∥xn-p∥2+αn∥f(xn)-p∥2+2λn(λnK2-β)∥Ax-Ay∥2.
Then, we obtain that
(3.12)2λn(β-λnK2)∥Ax-Ay∥2≤∥xn-p∥2-∥xn+1-p∥2+αn∥f(xn)-p∥2≤∥xn-xn+1∥(∥xn-p∥+∥xn+1-p∥)+αn∥f(xn)-p∥2.
By (C1), (C2), (C3), and (3.10), we get
(3.13)limn→∞∥Axn-Ap∥=0.
From Proposition 2.1 (ii) and Lemma 2.5, we also have
(3.14)∥un-p∥2=∥QC(xn-λnAxn)-QC(p-λnAp)∥2≤〈(xn-λnAxn)-(p-λnAp),J(un-p)〉=〈(xn-p)-λn(Axn-Ap),J(un-p)〉=〈xn-p,J(un-p)〉-λn〈Axn-Ap,J(un-p)〉≤12[∥xn-p∥2+∥un-p∥2-g∥xn-un∥]+λn∥Axn-Ap∥∥un-p∥.
So, we get,
(3.15)∥un-p∥2≤∥xn-p∥2-g∥xn-un∥+2λn∥Axn-Ap∥∥un-p∥.
Therefore, using (3.11), we obtain
(3.16)∥xn+1-p∥2≤βn∥xn-p∥2+(1-βn)[αn∥f(xn)-p∥2+(1-αn)∥un-p∥2]=βn∥xn-p∥2+αn(1-βn)∥f(xn)-p∥2+(1-αn)(1-βn)∥un-p∥2≤βn∥xn-p∥2+αn(1-βn)∥f(xn)-p∥2+(1-αn)(1-βn)[∥xn-p∥2-g(∥xn-un∥)+2λn∥Axn-Ap∥∥un-p∥]≤∥xn-p∥2+αn∥f(xn)-p∥2-g(∥xn-un∥)+2λn∥Axn-Ap∥∥un-p∥.
Then we get
(3.17)g(∥xn-un∥)≤∥xn-p∥2-∥xn+1-p∥2+αn∥f(xn)-p∥2+2λn∥Axn-Ap∥∥un-p∥≤∥xn-xn+1∥(∥xn-p∥+∥xn+1-p∥)+αn∥f(xn)-p∥2+2λn∥Axn-Ap∥∥un-p∥.
By (C1), (3.10), and (3.13), we have
(3.18)limn→∞g(∥xn-un∥)=0.
It follows from the property of g that
(3.19)limn→∞∥xn-un∥=0.
Again, we consider
(3.20)∥yn-p∥2=∥QC[αnf(xn)+(1-αn)un]-QCp∥2≤〈αn(f(xn)-p)+(1-αn)(un-p),J(yn-p)〉=αn〈f(xn)-p,J(yn-p)〉+(1-αn)〈un-p,J(yn-p)〉≤αn∥f(xn)-p∥∥yn-p∥+12[∥un-p∥2+∥yn-p∥2-g(∥un-yn∥)]≤αn∥f(xn)-p∥∥yn-p∥+12[∥xn-p∥2+∥yn-p∥2-g(∥un-yn∥)].
It follows that
(3.21)∥yn-p∥2≤2αn∥f(xn)-p∥∥yn-p∥+∥xn-p∥2-g(∥un-yn∥).
By using (3.11), we obtain
(3.22)∥xn+1-p∥2≤βn∥xn-p∥2+(1-βn)[αn∥f(xn)-p∥2+(1-αn)∥yn-p∥2]=βn∥xn-p∥2+αn(1-βn)∥f(xn)-p∥2+(1-αn)(1-βn)∥yn-p∥2≤βn∥xn-p∥2+αn(1-βn)∥f(xn)-p∥2+(1-αn)(1-βn)[2αn∥f(xn)-p∥∥yn-p∥+∥xn-p∥2-g(∥un-yn∥)]≤∥xn-p∥2+αn∥f(xn)-p∥2+2αn∥f(xn)-p∥∥yn-p∥-g(∥un-yn∥).
Therefore, we have
(3.23)g(∥un-yn∥)≤∥xn-p∥2-∥xn+1-p∥2+αn∥f(xn)-p∥2+2αn∥f(xn)-p∥∥yn-p∥≤∥xn-xn+1∥(∥xn-p∥+∥xn+1-p∥)+2αn∥f(xn)-p∥2+2αn∥f(xn)-un∥∥yn-un∥.
By (C1) and (3.10), we have
(3.24)limn→∞g(∥un-yn∥)=0.
From the property of g, we get
(3.25)limn→∞∥un-yn∥=0.
According (3.19) and (3.25), we also have
(3.26)limn→∞∥xn-yn∥=0.
Since
(3.27)∥xn-T(μn)xn∥≤∥xn-T(μn)yn∥+∥T(μn)yn-T(μn)xn∥≤∥xn-T(μn)yn∥+∥yn-xn∥,
from (3.9) and (3.26), we get
(3.28)limn→∞∥T(μn)xn-xn∥=0.
Now, we show that z∈F:=F(𝒮)∩VI(C,A). We can choose a sequence {xnk} of {xn} such that {xnk} is bounded, and there exists a subsequence {xnkj} of {xnk} which converges weakly to z. Without loss of generality, we can assume that xnk⇀z.
(I) We show that z∈VI(C,A). From the assumption, we see that control sequence {λnk} is bounded. So, there exists a subsequence {λnkj} that converges to λ0. We may assume, without loss of generality, that λnk⇀λ0. Observe that
(3.29)∥QC(xnk-λ0Axnk)-xnk∥≤∥QC(xnk-λ0Axnk)-ynk∥+∥ynk-xnk∥≤∥(xnk-λ0Axnk)-(xnk-λnkAxnk)∥+∥ynk-xnk∥≤M∥λnk-λ0∥+∥ynk-xnk∥,
where M is as appropriate constant such that M≥supn≥1{∥Axn∥}. It follows from (3.26) and λnk⇀λ0 that
(3.30)limk→∞∥QC(xnk-λ0Axnk)-xnk∥=0.
We know that QC(I-λ0A) is nonexpansive, and it follows from Lemma 2.3 that z∈F(QC(I-λ0A)). By using Lemma 2.14, we can obtain that z∈F(QC(I-λ0A))=VI(C,A).
(II) Next, we show that z∈F(𝒮). Let μnk≥0 such that
(3.31)μnk→0,∥T(μnk)xnk-xnk∥μnk→0,k→∞.
Fix t>0. Notice that
(3.32)∥xnk-T(t)z∥≤∑i=0[t/μnk]-1∥T((i+1)μnk)xnk-T(iμnk)xnk∥+∥T([tμnk]μk)xnk-T([tμnk]μnk)z∥+∥T([tμnk]μnk)z-T(t)z∥≤[tμnk]∥T(μnk)xnk-xnk∥+∥xnk-z∥+∥T(t-[tμnk]μnk)z-z∥≤t∥T(μnk)xnk-xnk∥μnk+∥xnk-p∥+∥T(t-[tμnk]μnk)z-z∥≤t∥T(μnk)xnk-xnk∥μnk+∥xnk-p∥+max{∥T(s)z-z∥:0≤s≤μnk}.
For all k∈ℕ, we have
(3.33)limsupk→∞∥xnk-T(t)z∥≤limsupk→∞∥xnk-z∥.
Since a Banach space E with a weakly sequentially continuous duality mapping satisfies the Opial’s condition, this implies that T(t)z=z. Therefore z∈F(𝒮), so z∈F.
Next, we show that limsupn→∞〈(f-I)x*,J(yn-x*)〉≤0, where x*=QFfx*, QF is a sunny nonexpansive retraction of C onto F. Since we have (3.26) and xnk⇀z, then we have ynk⇀z such that
(3.34)limsupn→∞〈(f-I)x*,J(yn-x*)〉=limk→∞〈(f-I)x*,J(ynk-x*)〉=limk→∞〈(f-I)x*,J(z-x*)〉≤0.
Finally, we show that {xn} converges strongly to x*. Suppose that {xn} does not converge strongly to x*, and then there exist ϵ>0 and a subsequence {xni} of {xn} such that ∥xni-x*∥>ϵ for all i∈ℕ. By Proposition 2.12, for this ϵ there exists r∈(0,1) such that
(3.35)∥f(xni)-f(p)∥≤r∥xni-p∥.
So, by Lemma 2.7, we have
(3.36)∥yni-x*∥2=∥QC[αnf(xni)+(1-αni)uni]-QCx*∥2≤〈αni(f(xni)-x*)+(1-αni)(uni-x*),J(yni-x*)〉=αni〈f(xni)-x*,J(yni-x*)〉+(1-αni)〈uni-x*,J(yni-x*)〉≤αni〈f(xni)-f(x*),J(yni-x*)〉+αni〈f(x*)-x*,J(yni-x*)〉+(1-αni)∥uni-x*∥∥yni-x*∥≤rαni∥xni-x*∥∥yni-x*∥+αni〈f(x*)-x*,J(yni-x*)〉+(1-αni)∥uni-x*∥∥yni-x*∥≤rαni∥xni-x*∥2+αni〈f(x*)-x*,J(yni-x*)〉+(1-αni)∥xni-x*∥2=(1-(1-r)αni)∥xni-x*∥2+αni〈f(x*)-x*,J(yni-x*)〉.
It follows from (3.11) that
(3.37)∥xni+1-p∥2≤βni∥xni-p∥2+(1-βni)∥yni-p∥2≤βni∥xni-p∥2+(1-βni)×[(1-(1-r)αni)∥xni-x*∥2+αni〈f(x*)-x*,J(yni-x*)〉]=[1-(1-r)(1-βni)αni]∥xni-x*∥2+(1-βni)αni〈f(x*)-x*,J(yni-x*)〉.
Now, from (C1), (3.34) and applying Lemma 2.9 to (3.37), we get ∥xn-x*∥→0 as n→∞. This is a contradiction, and hence the sequence {xn} converges strongly to x*∈F. The proof is completed.
Corollary 3.2.
Let E be a uniformly convex and 2-uniformly smooth Banach space with the best smooth constant K and C a nonempty closed convex subset of E. Let QC be a sunny nonexpansive retraction from E onto C and A:C→E be an β-inverse strongly accretive operator. Let 𝒮={T(t):t≥0} be a nonexpansive semigroup from C into it self and f be a Meir-Keeler contraction of C into itself. Suppose that F:=F(𝒮)∩VI(C,A)≠∅. For arbitrary given x1∈C, the sequences {xn} are generated by
(3.38)un=QC(xn-λnAxn),yn=QC[αnf(xn)+(1-αn)un],xn+1=βnxn+(1-βn)1tn∫0tnT(s)ynds,∀n≥1,
where {αn},{βn}⊂(0,1), and {λn}⊂(0,∞) satisfy the conditions (C1)–(C3) in Theorem 3.1 and assume that limn→∞supx∈C~∥(1/tn+1)∫0tn+1T(s)xds-(1/tn)∫0tnT(s)xds∥=0, C~ bounded subset of C, limn→∞μn=∞ and limn→∞(μn/μn+1)=1. Then {xn} converges strongly to x*=QFfx*∈F, which also solves the following variational inequality:
(3.39)〈(f-I)x*,J(z-x*)〉≤0,∀z∈F.
Corollary 3.3.
Let E be a uniformly convex and 2-uniformly smooth Banach space with the best smooth constant K and C a nonempty closed convex subset of E. Let QC be a sunny nonexpansive retraction from E onto C. Let 𝒮={T(t):t≥0} be a nonexpansive semigroup from C into itself and f be a Meir-Keeler contraction of C into itself. Suppose that F(𝒮)≠∅, {αn},{βn}⊂(0,1), and {μn}⊂(0,∞) satisfy the conditions (C1), (C2), (C4), and (C5) in Theorem 3.1. For arbitrary given x1∈C, the sequences {xn} are generated by
(3.40)yn=QC[αnf(xn)+(1-αn)xn],xn+1=βnxn+(1-βn)T(μn)yn,∀n≥1.
Then {xn} converges strongly to x*=QFfx*∈F(𝒮), which also solves the following variational inequality:
(3.41)〈(f-I)x*,J(z-x*)〉≤0,∀z∈F.
Proof.
Taking A=0 in Theorem 3.1, we can conclude the desired conclusion easily. This completes the proof.
Corollary 3.4.
Let E be a uniformly convex and 2-uniformly smooth Banach space with the best smooth constant K and C a nonempty closed convex subset of E. Let QC be a sunny nonexpansive retraction from E onto C and A:C→E be an β-inverse strongly accretive operator. Let f be a Meir-Keeler contraction of C into itself. Suppose that F:=VI(C,A)≠∅, {αn},{βn}⊂(0,1) and {λn}⊂(0,∞) satisfy the conditions (C1)–(C3) in Theorem 3.1. For arbitrary given x1∈C, the sequences {xn} are generated by
(3.42)un=QC(xn-λnAxn),yn=QC[αnf(xn)+(1-αn)un],xn+1=βnxn+(1-βn)yn,∀n≥1.
Then {xn} converges strongly to x*=QFfx*∈F, which also solves the following variational inequality:
(3.43)〈(f-I)x*,J(z-x*)〉≤0,∀z∈F.
Proof.
Taking μn=0 for all n≥1 in Theorem 3.1, we can conclude the desired conclusion easily. This completes the proof.
Corollary 3.5.
Let E be a uniformly convex and 2-uniformly smooth Banach space with the best smooth constant K and C a nonempty closed convex subset of E. Let QC be a sunny nonexpansive retraction from E onto C and A:C→E be an β strongly accretive and L-Lipschitz continuous operator. Let 𝒮={T(t):t≥0} be a nonexpansive semigroup from C into it self and f be a Meir-Keeler contraction of C into itself. Suppose that F:=F(𝒮)∩VI(C,A)≠∅, {αn},{βn}⊂(0,1) and {μn},{λn}⊂(0,∞) satisfy the conditions (C1) and (C3)–(C5) in Theorem 3.1. If the sequence {xn} is generated by x1∈C and (3.1) and limn→∞(λn+1-λn)=0 and 0<a≤λn≤b<β/K2L2, then the sequence {xn} converges strongly to x*=QFfx*∈F, which also solves the following variational inequality:
(3.44)〈(f-I)x*,J(z-x*)〉≤0,∀z∈F.
Proof.
Since A be an β strongly accretive and L-Lipschitz continuous operator of C into E, we have
(3.45)〈Ax-Ay,j(x-y)〉≥β∥x-y∥2≥βL2∥Ax-Ay∥2,∀x,y∈C.
Therefore, A is (β/L2)-inverse strongly accretive. Using Theorem 3.1, we can obtain that {xn} converges strongly to x*. This completes the proof.
The following corollary is defined in a real Hilbert space. Let C be a closed convex subset of a real Hilbert space H. Let A:C→H be a mapping. The classical variational inequality problems are to find x∈C such that
(3.46)〈Ax,y-x〉≥0,
for all y∈C. For every point x∈H, there exists a unique nearest point in C, denoted by PCx, such that
(3.47)∥x-PCx∥≤∥x-y∥,∀y∈C.PC is called the metric projection of H onto C. It is well known that PC is a nonexpansive mapping of H onto C and satisfies
(3.48)〈x-y,PCx-PCy〉≥∥PCx-PCy∥2,
for every x,y∈H. Moreover, PCx is characterized by the following properties: PCx∈C and
(3.49)〈x-PCx,y-PCx〉≤0,(3.50)∥x-y∥2≥∥x-PCx∥2+∥y-PCx∥2,
for all x∈H,y∈C.
It is well known in Hilbert spaces the smooth constant K=2/2 and J=I (identity mapping). From Theorem 3.1, we can obtain the following result immediately.
Corollary 3.6.
Let C be a nonempty compact convex subset of a real Hilbert space H. Let PC be a metric projection of H onto C and A:C→E be an β-inverse strongly accretive operator. Let 𝒮={T(t):t≥0} be a nonexpansive semigroup from C into itself and f be a Meir-Keeler contraction of C into itself. Suppose that F:=F(𝒮)∩VI(C,A)≠∅, {αn},{βn}⊂(0,1) and {μn},{λn}⊂(0,∞) satisfy the conditions (C1)–(C5) in Theorem 3.1. For arbitrary given x1∈C, the sequences {xn} are generated by
(3.51)un=PC(xn-λnAxn),yn=PC[αnf(xn)+(1-αn)un],xn+1=βnxn+(1-βn)T(μn)yn,∀n≥1.
Then {xn} converges strongly to x*=PFfx*∈F which also solves the following variational inequality:
(3.52)〈(f-I)x*,z-x*〉≤0,∀z∈F.
Remark 3.7.
Question and Open problems. Can we extend Theorem 3.1 to more general variational inequalities in the sense of Noor [1] on Banach spaces?
Acknowledgments
The first author gratefully acknowledges support provided by King Mongkuts University of Technology Thonburi (KMUTT) during the first authors stay at King Mongkuts University of Technology Thonburi (KMUTT) as a postdoctoral fellow. Moreover, the authors would like to thank the Higher Education Research Promotion and National Research University Project of Thailand, Office of the Higher Education Commission (NRU-CSEC no. 55000613) for financial support. Both authors thank the referees for their comments which improved the presentation of this paper.
NoorM. A.General variational inequalities19881211912210.1016/0893-9659(88)90054-7953368ZBL0655.49005A. NoorM.Some developments in general variational inequalities2004152119927710.1016/S0096-3003(03)00558-72050063ZBL1134.49304NoorM. A1975London, UKBrunel UniversityNoorM. A.Some iterative techniques for general monotone variational inequalities199946439140110.1080/023319399088444641776180ZBL0966.49010NoorM. A.New approximation schemes for general variational inequalities2000251121722910.1006/jmaa.2000.70421790406ZBL0964.49007NoorM. A.RassiasT. M.A class of projection methods for general variational inequalities2002268133434310.1006/jmaa.2001.78961893209ZBL1038.49017NoorM. A.NoorK. I.RassiasT. M.Some aspects of variational inequalities199347328531210.1016/0377-0427(93)90058-J1251844ZBL0788.65074YaoY.MarusterS.Strong convergence of an iterative algorithm for variational inequalities in Banach spaces2011541-232532910.1016/j.mcm.2011.02.0162801889ZBL1225.65067LauA. T. M.TheraM. A.BaillonJ. B.Amenability and fixed point property for semigroup of nonexpansive mappings1991252Harlow, UKLongman Scientific & Technical303313Pitman Research Notes in Mathematics Series1122837ZBL0789.47039LauA. T.-M.ZhangY.Fixed point properties od semigroups of non-expansive mappings200824525342554AoyamaK.IidukaH.TakahashiW.Weak convergence of an iterative sequence for accretive operators in Banach spaces200613353902235489ZBL1128.47056LiX. N.GuJ. S.Strong convergence of modified Ishikawa iteration for a nonexpansive semigroup in Banach spaces20107341085109210.1016/j.na.2010.04.0402653774ZBL1226.47081WangkeereeR.PreechasilpP.Modified noor iterations for nonexpansive semigroups with generalized contraction in Banach spaces20122012610.1186/1029-242X-2012-62891959GossezJ.-P.Lami DozoE.Some geometric properties related to the fixed point theory for nonexpansive mappings1972405655730310717ZBL0223.47025ReichS.Asymptotic behavior of contractions in Banach spaces19734415770032868910.1016/0022-247X(73)90024-3ZBL0275.47034KitaharaS.TakahashiW.Image recovery by convex combinations of sunny nonexpansive retractions1993223333421251942ZBL0815.47068BrowderF.E.Nonlinear operators and nonlinear equations of evolution in Banach space1976187881XuH. K.Inequalities in Banach spaces with applications199116121127113810.1016/0362-546X(91)90200-K1111623ZBL0757.46033KamimuraS.TakahashiW.Strong convergence of a proximal-type algorithm in a Banach space200213393894510.1137/S105262340139611X1972223LiuL. S.Ishikawa and Mann iterative process with errors for nonlinear strongly accretive mappings in Banach spaces1995194111412510.1006/jmaa.1995.12891353071ZBL0872.47031SuzukiT.Strong convergence of Krasnoselskii and Mann's type sequences for one-parameter nonexpansive semigroups without Bochner integrals2005305122723910.1016/j.jmaa.2004.11.0172128124ZBL1068.47085XuH.-K.Viscosity approximation methods for nonexpansive mappings2004298127929110.1016/j.jmaa.2004.04.0592086546ZBL1061.47060MeirA.KeelerE.A theorem on contraction mappings196928326329025029110.1016/0022-247X(69)90031-6ZBL0194.44904LimT.-C.On characterizations of Meir-Keeler contractive maps200146111312010.1016/S0362-546X(99)00448-41845580ZBL1009.54044