Abstract
The problem of finding a solution of a variational inequality over the set of common fixed points of a nonexpansive semigroup is considered in a real and uniformly convex Banach space without imposing the sequential weak continuity of the normalized duality mapping. Two new explicit iterative methods are introduced based on the steepest-descent method, and conditions are given to obtain their strong convergence. A numerical example is showed to illustrate the convergence analysis of the proposed methods.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Let H be a real Hilbert space with inner product \(\langle \cdot , \cdot \rangle \) and norm \(\Vert \cdot \Vert \). Let C be a nonempty, closed, convex subset of H, and let \(F:H\rightarrow H\) be a mapping. The classical variational inequality problem (VI(F, C) for short) is formulated as finding a point \(p^* \in C\) such that
This problem was introduced by Stampacchia [14]. Recently, it has been extended to study a large variety of methods arising in structural analysis, economics, optimization, operations research, and engineering sciences (see, for example, [6,7,8,9,10,11,12,13, 22] and the references therein).
In 2001, Yamada [19] proposed the hybrid steepest-descent method for solving problem VI(F, C) when the feasible set C is the set of common fixed points of a finite family of nonexpansive mappings to avoid the complexity caused by the use of the metric projection \(P_C\). Besides, these research works are important because they contain many applications arising from the theory of signal recovery problems, power control problems, bandwidth allocation problems, and optimal control problems (see [8,9,10] and the references therein). Based on the hybrid steepest-descent method introduced by Yamada, many authors have been considering methods for solving variational inequality over the feasible set C with more complicated structure such as the common fixed point set of a countably infinite family of nonexpansive mappings (see for example, Yao et al. [21] and Wang [17]) or nonexpansive semigroup which is an uncountably infinite family of nonexpansive mappings [20].
There have been studies showing that solutions of an evolution equation with a m-accretive mapping \(A: E \rightarrow E\) in a Banach space constitute a nonexpansive semigroup generated by operator A, and further, the set of common fixed points of \(\{T (s): s \ge 0 \}\) is the set of zero points of A, that is \(\mathscr {F}: =\cap _{s \ge 0}\)Fix\((T(s))=A^{-1}(0)\) (see [4] for more details).
Along with the results achieved on different methods for solving the variational inequality (1.1) in a Hilbert space H (for example, Buong and Duong [2], Yang et al. [20]), many authors have recently studied solution methods for variational inequalities in Banach spaces (to name a few, Suzuki [15], Thuy and Hieu [16]). It is known that, among Banach spaces, Hilbert space H is a space with very nice geometrical properties such as the parallelogram identity, the existence of an inner product, or the uniqueness of the projection onto a nonempty, closed and convex subset of H which might be not valid in a general Banach space.
It should be added that a number of mathematical problems and research are set in Banach spaces related to variational inequalities such as differential equations and partial differential equations in Banach spaces or fixed point problems in Banach spaces. Therefore, methods for solving variational inequalities in Banach spaces or extensions to Banach spaces of methods developed in Hilbert spaces have been a topical issue attracting the attention of mathematicians. For some recent published results on solution methods for variational inequalities in Banach spaces, one needs to assume, in order to ensure their strong convergence, the weak continuity of the normalized duality mapping (such as, Ceng et al. [3], Chen and Song [5], and Suzuki [15]). Until now it has been shown that the \(l^p, \ 1<p<\infty \), satisfies this weak continuity property, while the \(L^p{[a, b]}, \ 1<p<\infty \), does not (see [4]). A natural question arising here is whether it is possible to develop methods for solving variational inequalities in Banach spaces without requiring the weak continuity of the normalized duality mapping. If the answer is affirmative, then the scope of applications of the algorithms in question can be expanded towards more general Banach spaces such as \(L^p[a, b]\), rather than applicable only for \(l^p, \ 1<p<\infty \).
Now, consider problem (1.1) in a Banach space E with \(C=\mathscr {F}\), the set of common fixed points of a nonexpansive semigroup in E. Precisely, the problem has the form: Find an element \(p^* \in \mathscr {F} \) such that
where \(F: E \rightarrow E\) is a strongly accretive and strictly pseudocontractive mapping, \(j(p^*-p) \in J(p^*-p)\), \(J: E \rightarrow E^*\) being the normalized duality mapping of E, and \(\langle x, x^* \rangle \) denotes the value of \(x^* \in E^*\), the dual space of E, at the point \(x \in E\).
In order to solve (1.2), Thuy and Hieu [16] proposed recently three implicit iterative methods, when E is a uniformly convex Banach space with a uniformly Gâteaux differentiable norm. The first method is defined by
the second one is in the form
and finally the third one is formulated as
for \(k \ge 1\), where \(0 <t_k \rightarrow \infty \) as \(k \rightarrow \infty \) in (1.3) and (1.4), whereas \(0< v_k \rightarrow 0\) as \(k \rightarrow \infty \) in (1.5).
The following result has been proved when iteration (1.5) is used:
Theorem 1.1
[16] Let F be an \(\eta -\)strongly accretive and \(\gamma -\)strictly pseudocontractive mapping with \(\eta +\gamma >1\) and let \(\{T(s): \ s\ge 0\}\) be a nonexpansive semigroup on E, which is a real uniformly convex Banach space with a uniformly Gâteaux differentiable norm, such that \(\mathscr {F}=\cap _{s \ge 0}\mathrm{Fix}(T(s)) \ne \emptyset \). Then, the sequence \(\{y_k\}\), defined by (1.5) where \(\gamma _k \in (0,1), \ \lambda _k \in (0,1]\) and \(v_k >0\) such that \( \lim _{k \rightarrow \infty }v_k =\lim _{k \rightarrow \infty } \frac{\gamma _k}{v_k} = 0\), converges to a unique element \(p^*\), solving (1.2).
Based on (1.3), (1.4), and (1.5), we construct two explicit iteration methods that are strongly convergent to a solution of (1.2) without imposing a sequentially weakly continuity property on the normalized duality mapping of Banach spaces. The methods are defined by
and
where \(F_n\) and \(T_n\) are, respectively, defined by
and \(\{ \gamma _n \}\), \(\{ \lambda _n \}\), \(\{ t_n \}\) are three sequences satisfying the following conditions:
and
The remainder of the paper is organized as follows. In the next Section, some preliminary results are recalled. In Sect. 3, two explicit iteration methods are studied for solving a variational inequality problem over the set of common fixed points of a nonexpansive semigroup in the framework of Banach spaces. The strong convergence of the sequences generated by each of these algorithms is obtained. It is related to the strong convergence of implicit methods given in [16]. Section 4 is devoted to show a numerical example for illustration purpose of the theoretical result. A final conclusion of the paper is given in Sect. 5.
2 Preliminaries
For simplicity, the norms of E and \(E^*\) are denoted by the same symbol \(\Vert \cdot \Vert \). A mapping J from E into \(2^{E^*}\) satisfying the condition
is called a normalized duality mapping of E. It is well known that if \(x \ne 0\), then \(J(tx)=tJ(x)\) for all \(t >0\) and \(x \in E\), and \(J(-x)=-J(x)\).
A mapping \(T:E \rightarrow E\) is said to be a nonexpansive mapping on E if
and
denotes the fixed point set of T. Let \(\{T(s): s\ge 0\}\) be a nonexpansive semigroup on E, that is,
-
(1)
for each \(s>0\), T(s) is a nonexpansive mapping on E;
-
(2)
\(T(0)x = x\) for all \(x \in E\);
-
(3)
\(T(s_1+s_2) = T(s_1) \circ T(s_2)\) for all \(s_1, s_2\ge 0\);
-
(4)
for each \(x \in C\), the mapping \(T( \cdot )x\) from \([0,\infty )\) into E is continuous.
Denote \(\mathscr {F}:=\cap _{s\ge 0}\text {Fix}(T(s))\) the set of common fixed points of the nonexpansive semigroup \(\{T(s): s\ge 0\}\). Through this paper we assume that \(\mathscr {F} \ne \emptyset \). Concerning the unemptiness of the set \(\mathscr {F}\), some conditions posed on the nonexpansive semigroup \(\{T(s): s\ge 0\}\) were mentioned in [4], such as if \(\{T(s): s\ge 0\}\) is a bounded semigroup on a closed and convex subset C of a uniformly convex Banach space E, that is \(\sup _{s \ge 0} \Vert T(s) x \Vert < +\infty , \ \forall x \in C\), then \(\mathscr {F} \ne \emptyset \) and vice versa.
A mapping F with domain D(F) and range R(F) in E is called an \(\eta \)-strongly accretive and \(\gamma \)-strictly pseudocontractive mapping if it satisfies
and
respectively, for all x, \(y \in D(F)\) and some element \(j(x-y) \in J(x-y)\), where I denotes the identity mapping of E, and \(\eta > 0\), \(\gamma \in (0,1) \) are fixed constants. Clearly, from (2.1), it follows that \(\Vert Fx - Fy \Vert \le L \Vert x - y\Vert \) with \(L=1+1/ \gamma \). If \(L \in [0,1)\), then F is called contractive, and if F satisfies (2.1) with \(\gamma = 0\), then it is said to be pseudocontractive. Clearly, every nonexpansive mapping is continuous and pseudocontractive.
Let \(S_1(0):=\{ x \in E: \Vert x \Vert =1\}\) be the unit sphere in E. The space E is called to have a Gâteaux differentiable norm (or to be smooth) if the limit
exists for all x, y in the sphere \(S_1(0)\). The space E is said to have a uniformly Gâteaux differentiable norm if the limit is attained uniformly for \(x \in S_1(0)\).
It is well known that if E is smooth, then the normalized duality mapping J is single valued. And if the norm of E is uniformly Gâteaux differentiable, then J is norm to weak-star uniformly continuous on every bounded subset of E (see [4]). In the sequel, we shall denote the single valued normalized duality mapping by j.
Recall that a Banach space E is said to be strictly convex if, for \( x, y \in S_1(0) \) with \(x \ne y\), \(\Vert (1-\lambda )x+\lambda y\Vert <1\) for all \(\lambda \in (0,1)\), and uniformly convex if, for any \(\varepsilon \), \(0< \varepsilon \le 2\), the inequalities \(\Vert x\Vert \le 1\), \(\Vert y\Vert \le 1\) and \(\Vert x-y\Vert \ge \varepsilon \) imply that there exists a \(\delta =\delta (\varepsilon ) \ge 0\) such that \( \Vert (x+y)/2 \Vert \le 1- \delta \). It is well known that every uniformly convex Banach space is reflexive and strictly convex.
Now we give some facts that will be used in the proof of our results.
Lemma 2.1
[1] Let E be a real normed linear space and let J be the normalized duality mapping on E. Then, the following inequality holds
Lemma 2.2
[3] Let E be a real smooth Banach space and let \(F: E \rightarrow E\) be an \(\eta \)-strongly accretive and \(\gamma \)-strictly pseudocontractive mapping with \(\eta +\gamma >1\). Then, for any \(\lambda \in (0,1)\), \( I-\lambda F\) is a contraction with contractive coefficient \(1- \lambda \tau \), where \( \tau = 1-\sqrt{(1-\eta )/ \gamma } \in (0,1)\).
Lemma 2.3
[5] Let C be a nonempty, bounded, closed, convex subset of a uniformly convex Banach space E and let \(\{T(s): s \ge 0\}\) be a nonexpansive semigroup on C. Then, for any \(r >0\) and \(h > 0\),
where \(B_r=\{x \in E: \Vert x\Vert \le r\}\).
Lemma 2.4
[15] Let \(\{x_n\}\) and \(\{z_n\}\) be bounded sequences in a Banach space E such that \(x_{n+1} = (1- \gamma _n) x_n + \gamma _n z_n\) for \(n \ge 1\), where \(\{\gamma _n\} \subset (0,1)\) such that \(0<\liminf _{n \rightarrow \infty } \gamma _n \le \limsup _{n \rightarrow \infty } \gamma _n <1\). Assume that
Then \(\lim _{n\rightarrow \infty } \Vert x_n - z_n \Vert =0\).
Lemma 2.5
[18] Let \(\{a_n\}\), \(\{b_n\}\), and \(\{c_n\}\) be three sequences of nonnegative numbers such that
where \(b_n \in (0,1), \ \sum _{n=1}^\infty b_n = \infty \), \(\lim \limits _{n \rightarrow \infty } b_n = 0\), and \(\limsup \limits _{n \rightarrow \infty } c_n \le 0\). Then \(\lim \limits _{n \rightarrow \infty } a_n = 0\).
Proposition 2.1
Let E be a real uniformly convex Banach space with a uniformly Gâteaux differentiable norm and \(F: E \rightarrow E\) be an L-Lipschitz continuous mapping. Let \(\{T(t): t \ge 0\}\) be a nonexpansive semigroup on E. If there exists a bounded sequence \(\{x_n\}\) satisfying \(\lim _{n\rightarrow \infty }\Vert x_n-T(t)x_n\Vert = 0\) for all \(t \ge 0\) and if \(\lim _{k \rightarrow \infty }y_k = p^*\), where \(\{y_k\}\) is a sequence defined by (1.5), that is
then,
Proof
Let k and n be fixed integers. Then
and
So
Hence
By assumption, since \(v_k \ge 0\), we have
Consequently, the sequence \(\{x_n\}\) being bounded, it follows from (2.3) and (2.4) that
On the other hand, for all k and n, we have
Since F is Lipschitz continuous with constant L, since the sequences \(\{y_k\}\) and \(\{x_n\}\) are bounded, and \(\Vert y_k-p^*\Vert \rightarrow 0\), we obtain that
Then we can write
and
Hence, thanks to (2.5), we can deduce that
Since \(\varepsilon >0\) is arbitrary, we finally obtain that
This completes the proof. \(\square \)
3 Main Results
Theorem 3.1
Let E be a real uniformly convex Banach space with a uniformly Gâteaux differentiable norm and \(F: E \rightarrow E\) be an \(\eta \)-strongly accretive and \(\gamma \)-strictly pseudocontractive mapping with \(\eta + \gamma >1\). Let also \(\{T(t): t \ge 0\}\) be a nonexpansive semigroup on E such that \(\mathscr {F} = \cap _{t \ge 0}\mathrm{Fix} (T(t)) \ne \emptyset \). Then the sequence \(\{x_n\}\) defined by (1.6) converges strongly to the unique solution \(p^*\) of (1.2).
Proof
The proof consists of three steps.
Step 1. We show that there exists a positive constant \(M_1\) such that \(\Vert x_n \Vert , \Vert T_n x_n \Vert ,\)\( \Vert F_n x_n \Vert , \Vert F_{n+1} x_n - p\Vert , \Vert F x_n \Vert \le M_1\) for all \(n \ge 1\) and all \(p \in \mathscr {F}\).
Let \(n \ge 1\). For a fixed point \(p \in \mathscr {F}\), we have \(T_n p = p,\) and hence, by Lemma 2.2,
Therefore, the sequence \(\{x_n\}\) is bounded. Since \(T_n\) is nonexpansive and \(I- \lambda _n F\) is contractive, we have
and
Consequently the sequences \(\{T_nx_n\}\) and \(\{F_nx_n\}\) are bounded and thus also the sequence \(\{F_{n+1}x_n - p \}\). Since F is a \(\gamma \)-strictly pseudocontractive mapping the sequence \(\{Fx_n\}\) is also bounded. So, the existence of \(M_1\) is proved.
Step 2. We prove that \(\lim _{n\rightarrow \infty } \Vert T(t)x_n - x_n \Vert =0\), for all \(t \ge 0\).
Let \(n \ge 1\), and define \(z_n = T_n F_n x_n\). From (1.6), it follows that
and
Since
we obtain that
This together with (1.8) and (1.9) implies that
Hence, we can deduce from (1.10) and Lemma 2.4 that
Since \(\Vert F_nx_n - x_n \Vert \le \lambda _n M_1\) and \(\lambda _n \rightarrow 0\) as \(n \rightarrow \infty \), we have also
On the other hand, from (3.2), (3.3), and
we get
Now, for any \(t>0\), observe that
This together with (3.4) and Lemma 2.3 implies that
Step 3. We claim that \(\lim _{n\rightarrow \infty } \Vert x_n-p^*\Vert =0\), where \(p^*\) is the unique solution of the variational inequality (1.2).
By Theorem 1.1 (see also Theorem 3.2 in [16]), we have that \(p^*\) exists and is unique. Furthermore, we also know that the sequence \(\{y_k\}\) defined, for each k, by (1.5) converges to \(p^*\). So, by Proposition 2.1, we have that
Now, using the convexity of \(\Vert \cdot \Vert ^2\), Lemmas 2.1, 2.2, and the property \(j(-x)=-j(x)\) for all \(x \in E\), we can estimate the value of \(\Vert x_{n+1}-p^* \Vert ^2\) as follows:
or
where
Since \(\{ \lambda _n \}\) and \(\{ \gamma _n \}\) satisfy conditions (1.8) and (1.10), respectively, we have that
Using (3.6) combined with \(\lambda _n \rightarrow 0\) as \(n \rightarrow \infty \), we obtain that
Then, from Lemma 2.5 and (3.7) we can conclude that \(\lim _{n \rightarrow \infty } \Vert x_n - p^* \Vert ^2 = 0\). This completes the proof. \(\square \)
Theorem 3.2
Assume the conditions in Theorem 3.1 are satisfied. Then the sequence \(\{x_n\}\) defined by (1.7) converges strongly to the unique solution \(p^*\) of (1.2).
Proof
We prove the result in three steps.
Step 1. We show that the sequence \(\{x_n\}\) is bounded.
For a fixed point \(p \in \mathscr {F}\), we have \(T_n p = p\) and hence, by Lemma 2.2,
Therefore, \(\{x_n\}\) is bounded, and so are the sequences \(\{T_nx_n\}\), \(\{F_nx_n\}\), \(\{Fx_n\}\), and \(\{x_n - p \}\). We assume that they are bounded by a positive constant \(M_2\).
Step 2. We prove that \(\lim _{n\rightarrow \infty } \Vert T(t)x_n - x_n \Vert =0\) for all \(t \ge 0\).
Put \(z_n = T_nx_n-\gamma _n\lambda _nFx_n/(1-\gamma _n)\). From (1.7) it follows that
and
where \(\beta \) is a positive number in (0, 1) such that \(\gamma _n \le \beta \). This together with \(\lambda _n \rightarrow 0\) and \(\vert t_{n+1}-t_n \vert / t_{n+1} \rightarrow 0 \) as \( n \rightarrow \infty \) implies that
Hence, we deduce from (1.10) and Lemma 2.4 that
On the other hand, since
and \(\lambda _n \rightarrow 0\) as \(n \rightarrow \infty \), we have that \(\Vert z_n - T_n x_n \Vert \rightarrow 0\).
Finally, using the same argument as in the proof of Theorem 3.1, we obtain that \(\Vert x_n - T_n x_n \Vert \rightarrow 0\) and \(\Vert T(t)x_n - x_n \Vert \rightarrow 0\) for all \(t \ge 0\) as \(n \rightarrow \infty \).
Step 3. We claim that \(\lim _{n\rightarrow \infty } \Vert x_n-p^*\Vert =0\), where \(p^*\) is the unique solution of the variational inequality (1.2).
By Theorem 1.1 (see also Theorem 3.2 in [16]), we have that \(p^*\) exists and is unique. Furthermore, we also know that the sequence \(\{y_k\}\) defined, for each k, by (1.5) converges to \(p^*\). So, by Proposition 2.1, we have that
Now we estimate the value of \(\Vert x_{n+1}-p^* \Vert ^2\) as follows:
On the other hand, we can write
Since the operator F is \(\gamma \)-strictly pseudomonotone and \(\eta \)-strongly accretive, we have successively
Since \(\eta + \gamma >1\), we have that \(\frac{1-\eta }{\gamma }<1\). Hence, with \(\tau = 1-\sqrt{\frac{1-\eta }{\gamma }}\), we get
Consequently,
Hence,
That is, we have
Thus, using the property \(j(-x)=-j(x)\) for all \(x \in E\), we can deduce
or
where
Since \(\sum _{n=0}^\infty \lambda _n = \infty \), we have \( \sum _{k=0}^\infty d_n = \infty \). Using (3.8), we obtain \(\limsup _{n \rightarrow \infty } e_n \le 0\).
Then, from Lemma 2.5 and (3.9) we have that \(\lim _{n \rightarrow \infty } \Vert x_n - p^* \Vert ^2 = 0\). This completes the proof. \(\square \)
4 A Numerical Illustration
It is easy to see that our results are applicable for solving the following optimization problem: Find a point \(p^*\in C\) such that
Here the function \(\varphi : \mathbb {R}^N \rightarrow \mathbb {R}\) is assumed to have a strongly monotone and Lipschitz continuous derivative \(\varphi '(x)\) on the Euclidean space \( \mathbb {R}^N\), and C is the set of common fixed points of a nonexpansive semigroup \(\{T(t): t\ge 0\}\) on \(\mathbb {R}^N\).
For our computational purpose, we consider the case when \(N=500\), \(\varphi (x)=\Vert x-1\Vert ^2\), and \(C:=\mathscr {F}\), the fixed point set of a nonexpansive semigroup \(\{T(t): \mathbb {R}^{500} \rightarrow \mathbb {R}^{500}, \ t\ge 0\}, \ x \longmapsto T(t) x\) defined by the multiplication of the following matrix
by \((x_1, x_2, \ldots , x_{500})^\mathrm{T}\), where \(x =(x_1,x_2,\ldots , x_{500}) \in \mathbb {R}^{500}\) and \(\alpha , \beta \in \mathbb {R}\) are fixed constants. In this case, \(C:=\mathscr {F} = \{ x \in \mathbb {R}^{500}: \ x = (0,\ldots ,0,x_5,\ldots , x_{498},0,0)^\mathrm{T}\}\) is a closed and convex subset of \(\mathbb {R}^{500}\) and the optimal condition for (4.1) is the variational inequality (1.2) with \(F(x)=\nabla \varphi (x)=2(x-1)\). It is easy to see that
is the unique solution of (4.1).
We apply our methods (1.6) and (1.7) for solving (4.1) with \(F(x)=2(x-1)\), the derivative of \(\varphi \) at \(x \in \mathbb {R}^{500}\). Clearly, F is 2-Lipschitz continuous and 1-strongly monotone on \(\mathbb {R}^{500}\). For both methods, the initial guess is \(x_0=(5, 5,\ldots , 5)\in \mathbb {R}^{500}\). We choose \(\alpha = \pi /5, \beta =\pi /7\), and the other parameters are as follows: \(t_n =(n+1)^{2}\), \(\gamma _n=(n+1)^{-1/7}\), and \( \lambda _n =(n+1)^{-1/ 2 } \) in the computation of (1.6) and \(\gamma _n=(2n+1)^{-1/2}\), and \(\lambda _n =(5n+1)^{-1/ 3 } \) in (1.7).
The numerical results obtained with our example are given in Tables 1 and 2.
The above numerical results show that the methods considered in this paper, based on (1.6) and (1.7), are able to solve efficiently some class of optimization problems and variational inequalities with control conditions. However, it seems that method (1.7) is more efficient than method (1.6).
5 Conclusion
We studied a class of accretive variational inequalities, where the feasible set is the set of common fixed points of a nonexpansive semigroup in a real uniformly convex Banach space with a uniformly Gâteaux differentiable norm. We introduced two explicit iterative methods to solve the considered problem. Our convergence analyses guarantee that the methods converge to a solution under certain mild assumptions. In particular, we do not impose the use of a sequentially weakly continuity property on the normalized duality mapping of Banach spaces. We presented a numerical example to illustrate the convergence analyses of the proposed methods.
References
Agarwal, R.P., O’Regan, D., Sahu, D.R.: Fixed Point Theory for Lipschitzian-Type Mappings with Applications. Springer, Dordrecht (2009)
Buong, N., Duong, L.T.: An explicit iterative algorithm for a class of variational inequalities in Hilbert Spaces. J. Optim. Theory Appl. 151, 513–524 (2011)
Ceng, L.-C., Ansari, Q.H., Yao, J.-C.: Mann-type steepest-descent and modified steepest-descent methods for variational inequalities in Banach spaces. Numer. Funct. Anal. Optim. 29, 987–1033 (2008)
Cioranescu, I.: Geometry of Banach Spaces. Duality Mappings and Nonlinear Problems. Kluwer, Dordrecht (1990)
Chen, R., Song, Y.: Convergence to common fixed point of nonexpansive semigroup. J. Comput. Appl. Math. 200, 566–575 (2007)
Duvaut, G., Lions, J.L.: Inequalities in Mechanics and Physics. Springer, Berlin (1976)
Hlaváček, I., Haslinger, J., Nečas, J., Lovíšek, J.: Solution of Variational Inequalities in Mechanics. Springer, Berlin (1998)
Iiduka, H.: Fixed point optimization algorithm and its application to power control in CDMA data networks. Math. Program. 133, 227–242 (2012)
Iiduka, H.: Fixed point optimization algorithm and its application to network bandwidth allocation. J. Comput. Appl. Math. 236, 1733–1742 (2012)
Iiduka, H.: Fixed point optimization algorithms for distributed optimization in network systems. SIAM J. Optim. 23, 1–26 (2013)
Kinderlehrer, D., Stampacchia, G.: An Introduction to Variational Inequalities and Their Applications. Academic Press, New York (1980)
Lions, J.L., Stampacchia, G.: Variational inequalities. Commun. Pure Appl. Math. 20, 493–519 (1967)
Panagiotopoulos, P.D.: Inequality Problems in Mechanics and Applications—Convex and Nonconvex Energy Functions. Birkhäuser, Boston (1985)
Stampacchia, G.: Formes bilinéaires coercitives sur les ensembles convexes. C. R. Acad. Sci. Paris 258, 4413–4416 (1964)
Suzuki, T.: Strong convergence theorems for infinite families of nonexpansive mappings in general Banach spaces. Fixed Point Theory Appl. 2005(1), 103–123 (2005)
Thuy, N.T.T., Hieu, P.T.: Implicit iteration methods for variational inequalities in Banach spaces. Bull. Malays. Math. Sci. Soc. 36, 917–926 (2013)
Wang, S.: Convergence and weaker control conditions for hybrid iterative algorithms. Fixed Point Theory Appl. (2011). doi:10.1186/1687-1812-2011-3
Xu, H.-K.: Iterative algorithms for nonlinear operators. J. Lond. Math. Soc. 66, 240–256 (2002)
Yamada, I.: The hybrid steepest descent method for the variational inequality problem over the intersection of fixed point sets of nonexpansive mappings. In: Butnariu, D., Censor, Y., Reich, S. (eds.) Inherently Parallel Algorithms in Feasibility and Optimization and Their Applications, Chapter 8, pp. 473–504. Elsevier Science Publishers, Amsterdam (2001)
Yang, P., Yao, Y., Liou, Y.C., Chen, R.: Hybrid algorithms of nonexpansive semigroups for variational inequalities. J. Appl. Math. (2012). doi:10.1155/2012/634927
Yao, Y., Noor, M.A., Liou, Y.C.: A new hybrid iterative algorithm for variational inequalities. Appl. Math. Comput. 216, 822–829 (2010)
Zeidler, E.: Nonlinear Functional Analysis and Its Applications. Springer, New York (1985)
Acknowledgements
The authors are very grateful to the referees for their useful comments, which helped to improve the paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Poom Kumam.
Rights and permissions
About this article
Cite this article
Hieu, P.T., Thuy, N.T.T. & Strodiot, J.J. Explicit Iteration Methods for Solving Variational Inequalities in Banach Spaces. Bull. Malays. Math. Sci. Soc. 42, 467–483 (2019). https://doi.org/10.1007/s40840-017-0494-8
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40840-017-0494-8