# Strong convergence of hybrid Halpern iteration for Bregman totally quasi-asymptotically nonexpansive multi-valued mappings in reflexive Banach spaces with application

- Yi Li
^{1}Email author and - Hongbo Liu
^{1}

**2014**:186

https://doi.org/10.1186/1687-1812-2014-186

© Li and Liu; licensee Springer. 2014

**Received: **8 May 2014

**Accepted: **14 August 2014

**Published: **2 September 2014

## Abstract

In this paper, Bregman totally quasi-asymptotically nonexpansive multi-valued mappings in the framework of reflexive Banach spaces are established. Under suitable limit conditions, by using the shrinking projection method introduced by Takahashi, Kubota and Takeuchi, some strong convergence theorems for hybrid Halpern’s iteration for a countable family of Bregman totally quasi-asymptotically nonexpansive multi-valued mappings are proved. We apply our main results to solve classical equilibrium problems in the framework of reflexive Banach spaces. The main result presented in the paper improves and extends the corresponding result in the work by Chang (Appl. Math. Comput. 2013, doi:10.1016/j.amc.2013.11.074; Appl. Math. Comput. 228:38-48, 2014), Suthep (Comput. Math. Appl., 64:489-499, 2012), Yi Li (Fixed Point Theory Appl. 2013:197, 2013), Reich and Sabach (Nonlinear Anal. 73:122-135, 2010), Nilsrakoo and Saejung (Appl. Math. Comput. 217(14):6577-6586, 2011), Qin *et al*. (Appl. Math. Lett. 22:1051-1055, 2009), Wang *et al*. (J. Comput. Appl. Math. 235:2364-2371, 2011), Su *et al*. (Nonlinear Anal. 73:3890-3906, 2010) and others.

**MSC:**47J05, 47H09, 49J25.

## Keywords

## 1 Introduction

*D*be a nonempty and closed subset of a real Banach space

*X*. Let $N(D)$ and $CB(D)$ denote the family of nonempty subsets and nonempty, closed, and bounded subsets of

*D*, respectively. The

*Hausdorff metric*on $CB(D)$ is defined by

An element $p\in D$ is called *a fixed point* of the multi-valued mapping $T:D\to N(D)$ if $p\in T(p)$. The set of fixed points of *T* is denoted by $F(T)$.

where ${\alpha}_{n}\in (0,1)$. This method was introduced in 1967 by Halpern [10] and is often called Halpern’s iteration. In fact, he proved, in a real Hilbert space, the strong convergence of $\{{x}_{n}\}$ to a fixed point of the nonexpansive mapping *T*, where ${\alpha}_{n}={n}^{-a}$, $a\in (0,1)$.

Because of the simple construction, Halpern’s iteration is widely used to approximate fixed points of nonexpansive mappings and other classes of nonlinear mappings by mathematicians in different styles [3–42]. In particular, some strong convergence theorems for resolvents of accretive operators in Banach spaces were proved by Reich [11], and he also extended the result of Halpern from Hilbert spaces to uniformly smooth Banach spaces in [12]. In 2012, Halpern’s iteration for Bregman strongly nonexpansive mappings in reflexive Banach spaces was introduced and a strong convergence theorem for Bregman strongly nonexpansive mappings by Halpern’s iteration in the framework of reflexive Banach spaces was proved. Recently, a strong convergence theorem for Bregman strongly multi-valued nonexpansive mappings as regards Halpern’s iteration in the framework of reflexive Banach spaces was proved by Chang [1, 2], Suthep [3] and Li [4].

The purpose of our work is to introduce a modified Halpern iteration for a countable family of Bregman totally quasi-asymptotically nonexpansive multi-valued mappings in the framework of reflexive Banach spaces, and to prove strong convergence theorems for these iterations under suitable limit conditions by using the shrinking projection method. We use our results to solve equilibrium problems in the framework of reflexive Banach spaces. The main results presented in the paper improve and extend the corresponding results in the work by Chang [1, 2], Suthep [3], Li [4], and others.

## 2 Preliminaries

In this section, we recall some basic definitions and results which will be used in the following.

Let *X* be a real reflexive Banach space with a norm $\parallel \cdot \parallel $, and let ${X}^{\ast}$ be the dual space of *X*. Let $f:X\to (-\mathrm{\infty},+\mathrm{\infty}]$ be a proper, lower semi-continuous, and convex function. We denote by $domf=\{x\in X:f(x)<+\mathrm{\infty}\}$ the domain of *f*.

*f*at

*x*is the convex set defined by

*f*is the function ${f}^{\ast}:{X}^{\ast}\to (-\mathrm{\infty},+\mathrm{\infty}]$ defined by

Furthermore, equality holds if ${x}^{\ast}\in \partial f(x)$ (see [13]). The set ${lev}_{\le}^{f}(r):=\{x\in X:f(x)\le r\}$ for some $r\in \mathbb{R}$ is called a sublevel of *f*.

*f*on

*X*is called coercive [14], if the sublevel sets of

*f*are bounded, or equivalently,

*f*on

*X*is said to be

*strongly coercive*[15], if

*f*at

*x*in the direction

*y*is defined by

The function *f* is said to be *Gâteaux differentiable* at *x*, if ${lim}_{t\to {0}^{+}}\frac{f(x+ty)-f(x)}{t}$ exists for any *y*. In this case, ${f}^{\circ}(x,y)$ coincides with $\mathrm{\nabla}f(x)$, the value of the gradient $\mathrm{\nabla}f(x)$ of *f* at *x*. The function *f* is said to be *Gâteaux differentiable*, if it is Gâteaux differentiable for any $x\in intdomf$. The function *f* is said to be *Fréchet differentiable* at *x*, if this limit is attained uniformly in $\parallel y\parallel =1$. Finally, *f* is said to be *uniformly Fréchet differentiable* on a subset *D* of *X*, if the limit is attained uniformly, for $x\in D$ and $\parallel y\parallel =1$. It is well known that if *f* is Gâteaux differentiable (resp. Fréchet differentiable) on $intdomf$, then *f* is continuous and its Gâteaux derivative ∇*f* is norm-to-weak*, continuous (resp. continuous) on $intdomf$ (see [16, 17]).

**Definition 2.1** (*cf*. [18])

*f*is said to be

- (i)
essentially smooth, if

*∂f*is both locally bounded and single-valued on its domain; - (ii)
essentially strictly convex, if ${(\partial f)}^{-1}$ is locally bounded on its domain and

*f*is strictly convex on every convex subset of $dom\partial f$; - (iii)
Legendre, if it is both essentially smooth and essentially strictly convex.

**Remark 2.1** (*cf*. [19])

*X*be a reflexive Banach space. Then we have

- (a)
*f*is essentially smooth if and only if ${f}^{\ast}$ is essentially strictly convex; - (b)
${(\partial f)}^{-1}=\partial {f}^{\ast}$;

- (c)
*f*is Legendre if and only if ${f}^{\ast}$ is Legendre; - (d)
If

*f*is Legendre, then*∂f*is a bijection which satisfies $\mathrm{\nabla}f={(\mathrm{\nabla}{f}^{\ast})}^{-1}$, $ran\mathrm{\nabla}f=dom\mathrm{\nabla}{f}^{\ast}=intdom{f}^{\ast}$ and $ran\mathrm{\nabla}{f}^{\ast}=dom\mathrm{\nabla}f=intdomf$.

Examples of Legendre functions can be found in [30]. One important and interesting Legendre function is $\frac{1}{p}{\parallel \cdot \parallel}^{p}$ ($0<p<+\mathrm{\infty}$) when *X* is a smooth and strictly convex Banach space. In this case the gradient ∇*f* of *f* is coincident with the generalized duality mapping of *X*, *i.e.,* $\mathrm{\nabla}f={J}_{p}$. In particular, $\mathrm{\nabla}f=I$, the identity mapping in Hilbert spaces. In this paper, we always assume that *f* is Legendre.

The following crucial lemma was proved by Reich and Sabach [20].

**Lemma 2.1** (*cf*. [20])

*If* $f:X\to R$ *is uniformly Fréchet differentiable and bounded on bounded subsets of* *X*, *then* ∇*f* *is uniformly continuous on bounded subsets of* *X* *from the strong topology of* *X* *to the strong topology of* ${X}^{\ast}$.

is called the Bregman distance with respect to *f*.

*f*at $x\in intdomf$ is the function ${v}_{f}(x,t):[0,+\mathrm{\infty})\to [0,+\mathrm{\infty})$ defined by

*f*is called totally convex at

*x*, if ${v}_{f}(x,t)>0$ whenever $t>0$. The function

*f*is called totally convex, if it is totally convex at any point $x\in intdomf$, and it is said to be totally convex on bounded sets, if ${v}_{f}(B,t)>0$, for any nonempty bounded subset

*B*of and $t>0$, where the modulus of the total convexity of the function

*f*on the set

*B*is the function ${v}_{f}:intdomf\times [0,+\mathrm{\infty})\to [0,+\mathrm{\infty})$ defined by

We know that *f* is totally convex on bounded sets if and only if *f* is uniformly convex on bounded sets (*cf.* [22]).

*f*is said to be sequentially consistent [22], if for any two sequences $\{{x}_{n}\}$ and $\{{y}_{n}\}$ in

*X*such that the first sequence is bounded, the following implication holds:

Recall that the function *f* is called sequentially consistent, if for any two sequences $\{{x}_{n}\}$ and $\{{y}_{n}\}$ in $intdomf$ and dom*f*, respectively, and $\{{x}_{n}\}$ is bounded, ${D}_{f}({y}_{n},x)\to 0$, then $\parallel {y}_{n}-x\parallel \to 0$.

The following crucial lemma was proved by Butnariu and Iusem [23].

**Lemma 2.2** (*cf*. [23])

*If*$x\in intdomf$,

*then the following statements are equivalent*:

- (i)
*The function**f**is totally convex at**x*. - (ii)
*For any sequence*$\{{y}_{n}\}\subset domf$, ${D}_{f}({y}_{n},x)\to 0$,*then*$\parallel {y}_{n}-x\parallel \to 0$.

**Definition 2.2** (*cf*. [24])

Let *D* be a convex subset of $intdomf$ and let *T* be a multi-valued mapping of *D*. A point $p\in D$ is called an asymptotic fixed point of *T* if *D* contains a sequence $\{{x}_{n}\}$ which converges weakly to *p* such that $d({x}_{n},T{x}_{n})\to 0$ (as $n\to \mathrm{\infty}$).

We denote by $\stackrel{\u02c6}{F}(T)$ the set of asymptotic fixed points of *T*.

**Definition 2.3**A multi-valued mapping $T:D\to N(D)$ with a nonempty fixed point set is said to be:

- (i)Bregman strongly nonexpansive with respect to a nonempty $\stackrel{\u02c6}{F}(T)$, if${D}_{f}(p,z)\le {D}_{f}(p,x),\phantom{\rule{1em}{0ex}}\mathrm{\forall}x\in D,p\in \stackrel{\u02c6}{F}(T),z\in T(x)$
and if, whenever $\{{x}_{n}\}\subset D$ is bounded, $p\in \stackrel{\u02c6}{F}(T)$, and ${lim}_{n\to \mathrm{\infty}}[{D}_{f}(p,{x}_{n})-{D}_{f}(p,{z}_{n})]=0$, then ${lim}_{n\to \mathrm{\infty}}{D}_{f}({x}_{n},{z}_{n})=0$, where ${z}_{n}\in T{x}_{n}$.

- (ii)Bregman firmly nonexpansive if$\begin{array}{r}\u3008\mathrm{\nabla}f\left({x}^{\ast}\right)-\mathrm{\nabla}f\left({y}^{\ast}\right),{x}^{\ast}-{y}^{\ast}\u3009\\ \phantom{\rule{1em}{0ex}}\le \u3008\mathrm{\nabla}f(x)-\mathrm{\nabla}f(y),{x}^{\ast}-{y}^{\ast}\u3009,\phantom{\rule{1em}{0ex}}\mathrm{\forall}x,y\in D,{x}^{\ast}\in Tx,{y}^{\ast}\in Ty.\end{array}$
- (iii)Bregman quasi-asymptotically nonexpansive mapping with sequence $\{{k}_{n}\}\subset [1,+\mathrm{\infty})$, ${k}_{n}\to 1$ (as $n\to \mathrm{\infty}$), if $\stackrel{\u02c6}{F}(T)=F(T)\ne \mathrm{\varnothing}$ and${D}_{f}(p,z)\le {k}_{n}{D}_{f}(p,x),\phantom{\rule{1em}{0ex}}p\in F(T),\mathrm{\forall}z\in {T}^{n}x,x\in D.$
- (iv)Bregman totally quasi-asymptotically nonexpansive mapping with nonnegative real sequence $\{{v}_{n}\}$, $\{{\mu}_{n}\}$, ${v}_{n},{\mu}_{n}\to 0$ (as $n\to \mathrm{\infty}$) and a strictly increasing continuous function $\zeta :{R}^{+}\to {R}^{+}$ with $\zeta (0)=0$, if $\stackrel{\u02c6}{F}(T)=F(T)\ne \mathrm{\varnothing}$ and${D}_{f}(p,z)\le {D}_{f}(p,x)+{\nu}_{n}\zeta ({D}_{f}(p,x))+{\mu}_{n},\phantom{\rule{1em}{0ex}}p\in F(T),\mathrm{\forall}z\in {T}^{n}x,x\in D.$
- (v)
Closed, if for any sequence $\{{x}_{n}\}\subset D$ with ${x}_{n}\to x\in N(D)$ and $d(T{x}_{n},,y)\to 0$ ($y\in D$), then $y\in Tx$.

**Remark 2.2** (*cf*. [1])

From these definitions, it is obvious that if $\stackrel{\u02c6}{F}(T)=F(T)\ne \mathrm{\varnothing}$, then a Bregman strongly nonexpansive multi-valued mapping is a Bregman relatively nonexpansive mapping; a Bregman relatively nonexpansive multi-valued mapping is a Bregman quasi-nonexpansive multi-valued mapping; a Bregman quasi-nonexpansive multi-valued mapping is a Bregman quasi-asymptotically nonexpansive multi-valued mapping; a Bregman quasi-asymptotically nonexpansive multi-valued mapping must be a Bregman totally quasi-asymptotically nonexpansive multi-valued mapping. However, converses of these statements are not true.

In particular, the existence and approximation of Bregman firmly nonexpansive single value mappings was studied in [24]. It is also known that if *T* is Bregman firmly nonexpansive and *f* is Legendre function which is bounded, uniformly Fréchet differentiable, and totally convex on bounded subsets of *X*, then $F(T)=\stackrel{\u02c6}{F}(T)$ and $F(T)$ is closed and convex (*cf*. [24]). It also follows that every Bregman firmly nonexpansive mapping is Bregman strongly nonexpansive with respect to $F(T)=\stackrel{\u02c6}{F}(T)$. The class of single-valued Bregman totally quasi-asymptotically nonexpansive mappings was introduced first in [1]. For a wealth of results concerning this class of mappings (for example, see [1], Examples 2.11-2.15 and the references therein).

**Remark 2.3**Let

*X*be a uniformly smooth and uniformly convex Banach space, and

*D*is nonempty, closed, and convex subset. An operator $T:C\to N(D)$ is called a strongly relatively nonexpansive multi-valued mapping on

*X*, if $\stackrel{\u02c6}{F}(T)\ne \mathrm{\Phi}$ and

and, if whenever $\{{x}_{n}\}\subset D$ is bounded, $p\in \stackrel{\u02c6}{F}(T)$, and ${lim}_{n\to \mathrm{\infty}}[\varphi (p,{x}_{n})-\varphi (p,{z}_{n})]=0$, then ${lim}_{n\to \mathrm{\infty}}\varphi ({x}_{n},{z}_{n})=0$, where ${z}_{n}\in T{x}_{n}$ and $\varphi (x,y)={\parallel x\parallel}^{2}-2\u3008x,Jy\u3009+{\parallel y\parallel}^{2}$.

Now, we give an example of Bregman totally quasi-asymptotically nonexpansive multi-valued mapping.

**Example 2.1** (see [1], Example 2.11)

*D*be a unit ball in a real Hilbert space ${l}^{2}$, $f(x)={\parallel x\parallel}^{2}$. Since $\mathrm{\nabla}f(y)=2y$, the Bregman distance with respect to

*f*

where any $\{{a}_{i}\}$ is a sequence in $(0,1)$ such that ${\prod}_{i=2}^{\mathrm{\infty}}{a}_{i}=\frac{1}{2}$.

- (i)
$\parallel Tx-Ty\parallel \le 2\parallel x-y\parallel $, $\mathrm{\forall}x,y\in D$;

- (ii)
$\parallel {T}^{n}x-{T}^{n}y\parallel \le 2{\prod}_{j=2}^{n}{a}_{j}\parallel x-y\parallel $, $\mathrm{\forall}x,y\in D$, $n\ge 2$.

*D*is a unit ball in a real Hilbert space ${l}^{2}$, it follows from (2.2) that ${D}_{f}(x,y)={\parallel x-y\parallel}^{2}$, $\mathrm{\forall}x,y\in D$. Above inequality can be written as

This shows that the mapping *T* defined as above is a Bregman total quasi-asymptotically nonexpansive multi-valued mapping.

*D*be a nonempty, closed, and convex subset of

*X*. Let $f:X\to \mathbb{R}$ be a Gâteaux differentiable and totally convex function and $x\in X$. It is well known from [22] that $z={P}_{D}^{f}(x)$ if and only if

*f*, which is defined by

^{∗}lower semi-continuous and convex function (see [33]). Hence ${V}_{f}$ is convex in the second variable (see [26], Proposition 1(i), p.1047). Thus,

The properties of the Bregman projection and the relative projection operators were studied in [22] and [27].

In 2013, Yi Li and Jin-hua Zhu proved the following result, respectively.

Let *X* be a real reflexive Banach space and let $f:X\to (-\mathrm{\infty},+\mathrm{\infty}]$ be a strongly coercive Legendre function which is bounded, uniformly Fréchet differentiable, and totally convex on bounded subsets of *X*. Let *D* be a nonempty, closed, and convex subset of $intdomf$. ${\alpha}_{n}\in (0,1)$, ${lim}_{n\to \mathrm{\infty}}{\alpha}_{n}=0$, and $0<{lim\hspace{0.17em}inf}_{n\to \mathrm{\infty}}{\beta}_{n}\le {lim\hspace{0.17em}sup}_{n\to \mathrm{\infty}}{\beta}_{n}<1$.

*X*such that $F(T)=\stackrel{\u02c6}{F}(T)\ne \mathrm{\varnothing}$. Suppose that $u\in X$ and define the sequence $\{{x}_{n}\}$ by

Then $\{{x}_{n}\}$ strongly converges to ${P}_{F(T)}^{f}(u)$.

*X*such that $F(T)=\stackrel{\u02c6}{F}(T)\ne \mathrm{\varnothing}$. Suppose that $u\in X$ and define the sequence $\{{x}_{n}\}$ by

Then $\{{x}_{n}\}$ strongly converges to ${P}_{F(T)}^{f}(u)$.

In 2014, Chang SS proved the following result.

**Theorem 2.1** ([1])

*Let*

*X*

*be a real uniformly smooth*,

*uniformly convex*,

*and reflexive Banach space*,

*D*

*be a nonempty*,

*closed*,

*and convex subset of*

*X*.

*Let*$f:D\to (-\mathrm{\infty},+\mathrm{\infty}]$

*be a Legendre function which is bounded*,

*uniformly Fréchet differentiable*,

*and totally convex on bounded subsets of*

*D*

*and let*$\{{T}_{i}\}:D\to D$

*be a family of closed and uniformly Bregman total quasi*-

*asymptotically nonexpansive mappings with sequence*$\{{v}_{n}\}$, $\{{\mu}_{n}\}$, ${v}_{n},{\mu}_{n}\to 0$ (

*as*$n\to \mathrm{\infty}$)

*and let there be a strictly increasing continuous function*$\zeta :{R}^{+}\to {R}^{+}$

*with*$\zeta (0)=0$

*such that*,

*for each*$i\ge 1$, $\{{T}_{i}\}$

*is uniformly*${L}_{i}$-

*Lipschitz continuous*.

*Let*$\{{\alpha}_{n}\}$

*be a sequence in*[0,1]

*such that*${lim}_{n\to \mathrm{\infty}}{\alpha}_{n}=0$.

*Let*${x}_{n}$

*be a sequence generated by*

*where* ${\xi}_{n}={v}_{n}{sup}_{p\in \mathcal{F}}\zeta ({D}_{f}(p,{x}_{n}))+{\mu}_{n}$, $\mathcal{F}={\bigcap}_{i=1}^{\mathrm{\infty}}F({T}_{i})$, ${P}_{{D}_{n+1}}^{f}$ *is the Bregman projection of* *X* *onto* ${D}_{n+1}$. *If* ℱ *is nonempty and bounded*, *then* $\{{x}_{n}\}$ *converges strongly to* ${P}_{\mathcal{F}}^{f}{x}_{1}$.

**Definition 2.4**

- (1)A countable family of multi-valued mappings ${\{{T}_{i}:D\to N(D)\}}_{i=1}^{\mathrm{\infty}}$ is said to be uniformly Bregman totally quasi-asymptotically nonexpansive, if $\mathcal{F}={\bigcap}_{i=1}^{\mathrm{\infty}}F({T}_{i})\ne \mathrm{\varnothing}$ and there exist nonnegative real sequences $\{{v}_{n}\}$, $\{{\mu}_{n}\}$, ${v}_{n},{\mu}_{n}\to 0$ (as $n\to \mathrm{\infty}$) and a strictly increasing continuous function $\zeta :{R}^{+}\to {R}^{+}$ with $\zeta (0)=0$, such that${D}_{f}(p,{z}_{n,i})\le {D}_{f}(p,x)+{\nu}_{n}\zeta ({D}_{f}(p,x))+{\mu}_{n},\phantom{\rule{1em}{0ex}}p\in F(T),\mathrm{\forall}{z}_{n,i}\in {T}_{i}^{n}x,x\in D.$(2.6)
- (2)A countable family of multi-valued mappings ${\{{T}_{i}:D\to N(D)\}}_{i=1}^{\mathrm{\infty}}$ is said to be uniformly Bregman quasi-asymptotically nonexpansive, if $\mathcal{F}={\bigcap}_{i=1}^{\mathrm{\infty}}F({T}_{i})\ne \mathrm{\varnothing}$ and there exist nonnegative real sequences $\{{k}_{n}\}\in [1,+\mathrm{\infty})$, ${k}_{n}\to 1$ (as $n\to \mathrm{\infty}$), such that${D}_{f}(p,{z}_{n,i})\le {D}_{f}(p,x)+{\nu}_{n}\zeta ({D}_{f}(p,x))+{\mu}_{n},\phantom{\rule{1em}{0ex}}p\in F(T),\mathrm{\forall}{z}_{n,i}\in {T}_{i}^{n}x,x\in D.$(2.7)
- (3)A multi-valued mapping $T:D\to N(D)$ is said to be uniformly
*L*-Lipschitz continuous, if there exists a constant $L>0$ such that$H({T}^{n}x,{T}^{n}y)\le L\parallel x-y\parallel ,\phantom{\rule{1em}{0ex}}\mathrm{\forall}x,y\in D.$(2.8)

Now, we improve the above results, and the following main results are obtained.

## 3 Main results

To prove our main result, we first give the following propositions.

The proof of the following result in the case of single-valued Bregman totally quasi-asymptotically nonexpansive mappings was done in ([1], Lemma 2.6, and [24], Lemma 15.5). In the multi-valued case the proof is identical and therefore we will omit the exact details. The interesting reader will consult [1, 24].

**Proposition 3.1** *Let* $f:X\to (-\mathrm{\infty},+\mathrm{\infty}]$ *be a Legendre function and let* *D* *be a nonempty*, *closed*, *and convex subset of* $intdomf$. *Let* $T:D\to N(D)$ *be a Bregman totally quasi*-*asymptotically nonexpansive multi*-*valued mapping with respect to* *f*. *Then* $F(T)$ *is closed and convex*.

**Theorem 3.1**

*Let*

*X*

*be a real uniformly smooth*,

*uniformly convex*,

*and reflexive Banach space*,

*D*

*be a nonempty*,

*closed*,

*and convex subset of*

*X*.

*Let*$f:D\to (-\mathrm{\infty},+\mathrm{\infty}]$

*be a Legendre function which is bounded*,

*uniformly Fréchet differentiable*,

*and totally convex on bounded subsets of*

*D*

*and let*$\{{T}_{i}\}:D\to N(D)$

*be a family of closed and uniformly Bregman totally quasi*-

*asymptotically nonexpansive multi*-

*valued mappings with sequence*$\{{v}_{n}\}$, $\{{\mu}_{n}\}$, ${v}_{n},{\mu}_{n}\to 0$ (

*as*$n\to \mathrm{\infty}$),

*and let there be a strictly increasing continuous function*$\zeta :{R}^{+}\to {R}^{+}$

*with*$\zeta (0)=0$

*such that*,

*for each*$i\ge 1$, $\{{T}_{i}\}$

*are uniformly*${L}_{i}$-

*Lipschitz continuous*.

*Let*$\{{\alpha}_{n}\}$

*be a sequence in*$[0,1]$

*and*$\{{\beta}_{n}\}$

*be a sequence in*$(0,1)$

*satisfying the following conditions*:

- (i)
${lim}_{n\to \mathrm{\infty}}{\alpha}_{n}=0$;

- (ii)
$0<{lim}_{n\to \mathrm{\infty}}inf{\beta}_{n}\le {lim}_{n\to \mathrm{\infty}}sup{\beta}_{n}<1$.

*Let*${x}_{n}$

*be a sequence generated by*

*where* ${\xi}_{n}={v}_{n}{sup}_{p\in \mathcal{F}}\zeta ({D}_{f}(p,{x}_{n}))+{\mu}_{n}$, $\mathcal{F}={\bigcap}_{i=1}^{\mathrm{\infty}}F({T}_{i})$, ${P}_{{D}_{n+1}}^{f}$ *is the Bregman projection of* *X* *onto* ${D}_{n+1}$. *If* ℱ *is nonempty and bounded*, *then* $\{{x}_{n}\}$ *converges strongly to* ${P}_{\mathcal{F}}^{f}{x}_{1}$.

*Proof* (I) First, we prove that ℱ and ${D}_{n}$ are closed and convex subsets in *D*.

*D*. Therefore ℱ is a closed and convex subset in

*D*. We use the assumption that ${D}_{1}=D$ is closed and convex. Suppose that ${D}_{n}$ is closed and convex for some $n\ge 1$. In view of the definition of ${D}_{f}(\cdot ,\cdot )$, we have

This shows that ${D}_{n+1}$ is closed and convex. The conclusions are proved.

(II) Next, we prove that $\mathcal{F}\subset {D}_{n}$, for all $n\ge 1$.

In fact, it is obvious that $\mathcal{F}\subset {D}_{1}$. Suppose that $\mathcal{F}\subset {D}_{n}$ for some $n\ge 1$.

- (III)
Now we prove that $\{{x}_{n}\}$ converges strongly to some point ${p}^{\ast}$.

Therefore $\{{D}_{f}({x}_{n},{x}_{1})\}$ is bounded, and so is $\{{x}_{n}\}$. Since ${x}_{n}={P}_{{D}_{n}}^{f}{x}_{1}$ and ${x}_{n+1}={P}_{{D}_{n+1}}^{f}{x}_{1}\in {D}_{n+1}\subset {D}_{n}$, we have ${D}_{f}({x}_{n},{x}_{1})\le {D}_{f}({x}_{n+1},{x}_{1})$. This implies that $\{{D}_{f}({x}_{n},{x}_{1})\}$ is nondecreasing. Hence ${lim}_{n\to \mathrm{\infty}}{D}_{f}({x}_{n},{x}_{1})$ exists.

It follows from Lemma 2.2 that ${lim}_{n\to \mathrm{\infty}}\parallel {x}_{m}-{x}_{n}\parallel =0$. Hence $\{{x}_{n}\}$is a Cauchy sequence in *D*. Since *D* is complete, without loss of generality, we can assume that ${lim}_{n\to \mathrm{\infty}}{x}_{n}={p}^{\ast}$ (some point in *D*).

- (IV)
Now we prove that ${p}^{\ast}\in \mathcal{F}$.

This implies that $\{{w}_{n,i}\}$ is uniformly bounded.

and this implies that $\{{\omega}_{n,i}\}$ is also uniformly bounded.

for each $i\ge 1$.

*f*is uniformly continuous on each bounded subset of

*X*, we have

*J*is uniformly continuous, this shows that

for each $i\ge 1$.

We get ${lim}_{n\to \mathrm{\infty}}\parallel H({T}_{i}^{n+1}{x}_{n})-H({T}_{i}^{n}{x}_{n})\parallel =0$. Since ${lim}_{n\to \mathrm{\infty}}{w}_{n,i}={p}^{\ast}$ and ${lim}_{n\to \mathrm{\infty}}{x}_{n}={p}^{\ast}$, we have ${lim}_{n\to \mathrm{\infty}}d(H({T}_{i}{T}_{i}^{n}{x}_{n}),{p}^{\ast})=0$.

*i.e.*, for each $i\ge 1$, ${p}^{\ast}\in F({T}_{i})$. By the arbitrariness of $i\ge 1$, we have ${p}^{\ast}\in \mathcal{F}$.

- (V)
Finally we prove that ${p}^{\ast}={P}_{\mathcal{F}}^{f}{x}_{1}$ and so ${x}_{n}\to {P}_{\mathcal{F}}^{f}{x}_{1}={p}^{\ast}$.

which yields ${p}^{\ast}=u={P}_{\mathcal{F}}^{f}{x}_{1}$. Therefore, ${x}_{n}\to {P}_{\mathcal{F}}^{f}{x}_{1}$. The proof of Theorem 3.1 is completed. □

By Remark 2.2, the following corollary is obtained.

**Theorem 3.2**

*Let*

*D*,

*X*, $\{{\alpha}_{n}\}$, $\{{\beta}_{n}\}$,

*and*

*f*

*be the same as in Theorem*3.1,

*Let*$\{{T}_{i}\}:D\to N(D)$

*be a family of closed and uniformly Bregman quasi*-

*asymptotically nonexpansive multi*-

*valued mappings with sequence*$\{{k}_{n}\}\subset [1,+\mathrm{\infty})$, ${k}_{n}\to 1$ (

*as*$n\to \mathrm{\infty}$)

*such that*,

*for each*$i\ge 1$, $\{{T}_{i}\}$

*be uniformly*${L}_{i}$-

*Lipschitz continuous*.

*Let*$\{{\alpha}_{n}\}$

*be a sequence in*$[0,1]$

*and*$\{{\beta}_{n}\}$

*be a sequence in*$(0,1)$

*satisfying the following conditions*:

- (i)
${lim}_{n\to \mathrm{\infty}}{\alpha}_{n}=0$;

- (ii)
$0<{lim}_{n\to \mathrm{\infty}}inf{\beta}_{n}\le {lim}_{n\to \mathrm{\infty}}sup{\beta}_{n}<1$.

*Let*${x}_{n}$

*be a sequence generated by*

*where* ${\xi}_{n}=({k}_{n}-1){sup}_{p\in \mathcal{F}}\zeta ({D}_{f}(p,{x}_{n}))$, $\mathcal{F}={\bigcap}_{i=1}^{\mathrm{\infty}}F({T}_{i})$, ${P}_{{D}_{n+1}}^{f}$ *is the Bregman projection of* *X* *onto* ${D}_{n+1}$. *If* ℱ *is nonempty*, *then* $\{{x}_{n}\}$ *converges strongly to* ${P}_{\mathcal{F}}^{f}{x}_{1}$.

As a direct consequence of Theorem 3.1 and Remark 2.3, we obtain the convergence result concerning Bregman totally quasi-asymptotically nonexpansive multi-valued mappings in a uniformly smooth and uniformly convex Banach space.

**Theorem 3.3** *Let* *X* *be a uniformly smooth and uniformly convex Banach space and* $J:X\to {2}^{{X}^{\ast}}$ *is the normalized duality mapping*. *Let* *D* *be a nonempty*, *closed*, *and convex subset on* *X* *and let* $T:D\to N(D)$ *be a family of closed and uniformly Bregman totally quasi*-*asymptotically nonexpansive multi*-*valued mappings with sequence* $\{{v}_{n}\}$, $\{{\mu}_{n}\}$, ${v}_{n},{\mu}_{n}\to 0$ (*as* $n\to \mathrm{\infty}$) *and a strictly increasing continuous function* $\zeta :{R}^{+}\to {R}^{+}$ *with* $\zeta (0)=0$ *such that*, *for each* $i\ge 1$, $\{{T}_{i}\}$ *be uniformly* ${L}_{i}$-*Lipschitz continuous*. *Let* $\{{\alpha}_{n}\}$ *be a sequence in* $[0,1]$ *and* $\{{\beta}_{n}\}$ *be a sequence in* $(0,1)$ *satisfying the following conditions*:

(C1) ${lim}_{n\to \mathrm{\infty}}{\alpha}_{n}=0$ *and* ${\sum}_{n=1}^{\mathrm{\infty}}{\alpha}_{n}=\mathrm{\infty}$,

(C2) $0<{lim\hspace{0.17em}inf}_{n\to \mathrm{\infty}}{\beta}_{n}\le {lim\hspace{0.17em}sup}_{n\to \mathrm{\infty}}{\beta}_{n}<1$.

*Let*${x}_{n}$

*be a sequence generated by*

*where* ${\xi}_{n}={\nu}_{n}{sup}_{p\in \mathcal{F}}\zeta (\varphi (p,{x}_{n}))+{\mu}_{n}$, $\mathcal{F}={\bigcap}_{i=1}^{\mathrm{\infty}}F({T}_{i})$, ${\mathrm{\Pi}}_{{D}_{n+1}}$ *is a projection of* *X* *onto* ${D}_{n+1}$. *If* ℱ *is nonempty and bounded*, *then* $\{{x}_{n}\}$ *converges strongly to* ${\mathrm{\Pi}}_{\mathcal{F}}{x}_{1}$.

Now, we provide examples of multi-valued mappings to which the results of the paper can be applied.

**Example 3.1**Let

*D*be a unit ball in a real Hilbert space ${l}^{2}$, $f(x)={\parallel x\parallel}^{2}$. Since $\mathrm{\nabla}f(y)=2y$, the Bregman distance with respect to

*f*

where any ${\{{a}_{j}\}}_{j=1}^{\mathrm{\infty}}$ is a sequence in $(0,1)$ such that ${\prod}_{j=2}^{\mathrm{\infty}}{a}_{j}=\frac{1}{2}$.

From Example 2.1, we know that ${\{{T}_{i}\}}_{i=1}^{\mathrm{\infty}}$ is a family of closed and uniformly Bregman totally quasi-asymptotically nonexpansive multi-valued mappings with sequence $\{{v}_{n}\}$, $\{{\mu}_{n}\}$, ${v}_{n},{\mu}_{n}\to 0$ (as $n\to \mathrm{\infty}$) and a strictly increasing continuous function $\zeta :{R}^{+}\to {R}^{+}$ with $\zeta (0)=0$ such that, for each $i\ge 1$, ${\{{T}_{i}\}}_{i=1}^{\mathrm{\infty}}$ is uniformly ${L}_{i}$-Lipschitz continuous. $\{{\alpha}_{n}\}$, $\{{\beta}_{n}\}$ and *f* are the same as in Theorem 3.1. Let $\{{x}_{n}\}$ be a sequence generated by (3.1), then $\{{x}_{n}\}$ converges strongly to ${P}_{\mathcal{F}}^{f}{x}_{1}$, where $\mathcal{F}={\bigcap}_{i=1}^{\mathrm{\infty}}F({T}_{i})$ is nonempty, ${P}_{\mathcal{F}}^{f}$ is the Bregman projection of *X* onto ℱ.

## 4 Application

In order to emphasize the importance of Theorem 3.1, we illustrate an application with the following important example, which entails equilibrium problems in the framework of reflexive Banach spaces.

Let *X* be a smooth, strictly convex, and reflexive Banach space, let *D* be a nonempty, closed, and convex subset of *X* and let $G:D\times D\to R$ be a bifunction satisfying the conditions: (A1) $G(x,x)=0$, for all $x\in D$; (A2) $G(x,y)+G(y,x)\le 0$, for any $x,y\in D$; (A3) for each $x,y,z\in D$, ${lim}_{t\to 0}G(tz+(1-t)x,y)\le G(x,y)$; (A4) for each given $x\in D$, the function $y\mapsto f(x,y)$ is convex and lower semicontinuous. The ‘so-called’ *equilibrium problem* for *G* is to find a ${x}^{\ast}\in D$ such that $G({x}^{\ast},y)\ge 0$, for each $y\in D$. The set of its solutions is denoted by $EP(G)$.

*G*[5] is the operator $Re{s}_{G}^{f}:X\to {2}^{D}$ defined by

*G*satisfies conditions (A1)-(A4), then $dom(Re{s}_{G}^{f})=X$ (see [5]). We also know:

- (1)
$Re{s}_{G}^{f}$ is single-valued;

- (2)
$Re{s}_{G}^{f}$ is a Bregman firmly nonexpansive mapping, so it is a closed Bregman total quasi-asymptotically nonexpansive mapping;

- (3)
$F(Re{s}_{G}^{f})=EP(G)$;

- (4)
$EP(G)$ is a closed and convex subset of

*D*; - (5)for all $x\in X$ and for all $p\in F(Re{s}_{G}^{f})$, we have${D}_{f}(p,Re{s}_{G}^{f}(x))+{D}_{f}(Re{s}_{G}^{f}(x),x)\le {D}_{f}(p,x).$(4.2)

In addition, by Reich and Sabach [24], if *f* is uniformly Fréchet differentiable and bounded on bounded subsets of *X*, then we see that $F(Re{s}_{G}^{f})=\stackrel{\u02c6}{F}(Re{s}_{G}^{f})=EP(G)$ is closed and convex. Hence, by replacing $T=Re{s}_{G}^{f}$ in Theorem 3.1, we obtain the following result.

**Theorem 4.1**

*Let*

*D*,

*X*, $\{{\alpha}_{n}\}$, $\{{\beta}_{n}\}$,

*and*

*f*

*be the same as in Theorem*3.1.

*Let*$\{{G}_{i}:D\times D\to R\}$

*be a countable families of bifunction which satisfies the conditions*(A1)-(A4)

*such that*$EP({G}_{i})\ne \mathrm{\varnothing}$.

*Let*$Re{s}_{{G}_{i}}^{f}(x):D\to {2}^{D}$, $i=1,2,\dots $ ,

*be the family of mappings defined by*

*Let the sequence*$\{{x}_{n}\}$

*be defined by*

*where* ${\xi}_{n}=({k}_{n}-1){sup}_{p\in \mathcal{F}}\zeta ({D}_{f}(p,{x}_{n}))$, $\mathcal{F}={\bigcap}_{i=1}^{\mathrm{\infty}}F(Re{s}_{{G}_{i}}^{f})$, ${P}_{{D}_{n+1}}^{f}$ *is the Bregman projection of* *X* *onto* ${D}_{n+1}$. *If* ℱ *is nonempty*, *then* $\{{x}_{n}\}$ *converges strongly to* ${P}_{\mathcal{F}}^{f}{x}_{1}$, *which is a common solution of the system of equilibrium problems for* ${G}_{m}$, $m=1,2,\dots $ .

## Declarations

### Acknowledgements

The authors are very grateful to both reviewers for careful reading of this paper and for their comments.

## Authors’ Affiliations

## References

- Chang SS, Wang L, Wang XR, Chan CK: Strong convergence theorems for Bregman totally quasi-asymptotically nonexpansive mappings in reflexive Banach spaces.
*Appl. Math. Comput.*2013. 10.1016/j.amc.2013.11.074Google Scholar - Chang SS, Wang L: Strong convergence theorems for Bregman totally quasi-asymptotically nonexpansive mappings in reflexive Banach spaces.
*Appl. Math. Comput.*2014, 228: 38–48.View ArticleMathSciNetGoogle Scholar - Suthep S, Yeol JC, Prasit C: Halpern’s iteration for Bregman strongly nonexpansive mappings in reflexive Banach spaces.
*Comput. Math. Appl.*2012, 64: 489–499. 10.1016/j.camwa.2011.12.026View ArticleMathSciNetGoogle Scholar - Li Y, Liu HB, Zheng K: Halpern’s iteration for Bregman strongly nonexpansive multi-valued mappings in reflexive Banach spaces with applications.
*Fixed Point Theory Appl.*2013., 2013: Article ID 197Google Scholar - Reich S, Sabach S: Two strong convergence theorems for Bregman strongly nonexpansive operators in reflexive Banach spaces.
*Nonlinear Anal.*2010, 73: 122–135. 10.1016/j.na.2010.03.005View ArticleMathSciNetGoogle Scholar - Nilsrakoo W, Saejung S: Strong convergence theorems by Halpern-Mann iterations for relatively nonexpansive mappings in Banach spaces.
*Appl. Math. Comput.*2011, 217(14):6577–6586. 10.1016/j.amc.2011.01.040View ArticleMathSciNetGoogle Scholar - Qin XL, Cho YJ, Kang SM, Zhou HY: Convergence of a modified Halpern-type iterative algorithm for quasi-
*ϕ*-nonexpansive mappings.*Appl. Math. Lett.*2009, 22: 1051–1055. 10.1016/j.aml.2009.01.015View ArticleMathSciNetGoogle Scholar - Wang ZM, Su YF, Wang DX, Dong YC: A modified Halpern-type iteration algorithm for a family of hemi-relative nonexpansive mappings and systems of equilibrium problems in Banach spaces.
*J. Comput. Appl. Math.*2011, 235: 2364–2371. 10.1016/j.cam.2010.10.036View ArticleMathSciNetGoogle Scholar - Su YF, Xu HK, Zhang X: Strong convergence theorems for two countable families of weak relatively nonexpansive mappings and applications.
*Nonlinear Anal.*2010, 73: 3890–3906. 10.1016/j.na.2010.08.021View ArticleMathSciNetGoogle Scholar - Halpern B: Fixed points of nonexpanding maps.
*Bull. Am. Math. Soc.*1967, 73: 957–961. 10.1090/S0002-9904-1967-11864-0View ArticleGoogle Scholar - Reich S: Strong convergence theorems for resolvents of accretive operators in Banach spaces.
*J. Math. Anal. Appl.*1980, 75: 287–292. 10.1016/0022-247X(80)90323-6View ArticleMathSciNetGoogle Scholar - Reich S: Approximating fixed points of nonexpansive mappings.
*Panam. Math. J.*1994, 4: 23–28.Google Scholar - Rockafellar RT: Level sets and continuity of conjugate convex functions.
*Trans. Am. Math. Soc.*1966, 123: 46–63. 10.1090/S0002-9947-1966-0192318-XView ArticleMathSciNetGoogle Scholar - Hiriart-Urruty JB, Lemaréchal C Grundlehren der mathematischen Wissenschaften 306. In
*Convex Analysis and Minimization Algorithms II: Advanced Theory and Bundle Methods*. Springer, Berlin; 1993.View ArticleGoogle Scholar - Zǎlinescu C:
*Convex Analysis in General Vector Spaces*. World Scientific, River Edge; 2002.View ArticleGoogle Scholar - Asplund E, Rockafellar RT: Gradients of convex functions.
*Trans. Am. Math. Soc.*1969, 139: 443–467.View ArticleMathSciNetGoogle Scholar - Bonnans JF, Shapiro A:
*Perturbation Analysis of Optimization Problems*. Springer, New York; 2000.View ArticleGoogle Scholar - Bauschke HH, Borwein JM, Combettes PL: Essential smoothness, essential strict convexity and Legendre functions in Banach spaces.
*Commun. Contemp. Math.*2001, 3: 615–647. 10.1142/S0219199701000524View ArticleMathSciNetGoogle Scholar - Bauschke HH, Borwein JM, Combettes PL: Bregman monotone optimization algorithms.
*SIAM J. Control Optim.*2003, 42: 596–636. 10.1137/S0363012902407120View ArticleMathSciNetGoogle Scholar - Reich S, Sabach S: A strong convergence theorem for a proximal-type algorithm in reflexive Banach spaces.
*J. Nonlinear Convex Anal.*2009, 10: 471–485.MathSciNetGoogle Scholar - Bregman LM: The relaxation method for finding the common point of convex sets and its application to the solution of problems in convex programming.
*USSR Comput. Math. Math. Phys.*1967, 7: 200–217.View ArticleGoogle Scholar - Butnariu D, Resmerita E: Bregman distances, totally convex functions and a method for solving operator equations in Banach spaces.
*Abstr. Appl. Anal.*2006., 2006: Article ID 84919Google Scholar - Butnariu D, Iusem AN:
*Totally Convex Functions for Fixed Points Computation and Infinite Dimensional Optimization*. Kluwer Academic, Dordrecht; 2000.View ArticleGoogle Scholar - Reich S, Sabach S: Existence and approximation of fixed points of Bregman firmly nonexpansive mappings in reflexive Banach spaces. In
*Fixed-Point Algorithms for Inverse Problems in Science and Engineering*. Springer, New York; 2011.Google Scholar - Goebel K, Kirk WA: A fixed point theorem for asymptotically nonexpansive mappings.
*Proc. Am. Math. Soc.*1972, 35: 171–174. 10.1090/S0002-9939-1972-0298500-3View ArticleMathSciNetGoogle Scholar - Martín-Márquez V, Reich S, Sabach S: Iterative methods for approximating fixed points of Bregman nonexpansive operators.
*Discrete Contin. Dyn. Syst.*2013, 6: 1043–1063.View ArticleGoogle Scholar - Kohsaka F, Takahashi W: Proximal point algorithms with Bregman functions in Banach spaces.
*J. Nonlinear Convex Anal.*2005, 6: 505–523.MathSciNetGoogle Scholar - Zhu JH, Chang SS: Halpern-Mann’s iterations for Bregman strongly nonexpansive mappings in reflexive Banach spaces.
*J. Inequal. Appl.*2013., 2013: Article ID 146Google Scholar - Rockafellar RT:
*Convex Analysis*. Princeton University Press, Princeton; 1970.View ArticleGoogle Scholar - Bauschke HH, Borwein JM: Legendre functions and the method of random Bregman projections.
*J. Convex Anal.*1997, 4: 27–67.MathSciNetGoogle Scholar - Alber YI: Metric and generalized projection operators in Banach spaces: properties and applications. In
*Theory and Applications of Nonlinear Operator of Accretive and Monotone Type*. Edited by: Kartsatos AG. Dekker, New York; 1996.Google Scholar - Censor Y, Lent A: An iterative row-action method for interval convex programming.
*J. Optim. Theory Appl.*1981, 34: 321–353. 10.1007/BF00934676View ArticleMathSciNetGoogle Scholar - Phelps RP Lecture Notes in Mathematics. In
*Convex Functions, Monotone Operators, and Differentiability*. 2nd edition. Springer, Berlin; 1993.Google Scholar - Li C, Song W, Yao JC: The Bregman distance approximate compactness and convexity of Chebychev sets in Banach spaces.
*J. Approx. Theory*2010, 162: 1128–1149. 10.1016/j.jat.2009.12.006View ArticleMathSciNetGoogle Scholar - Maingé PE: The viscosity approximation process for quasi-nonexpansive mappings in Hilbert spaces.
*Comput. Math. Appl.*2010, 59: 74–79. 10.1016/j.camwa.2009.09.003View ArticleMathSciNetGoogle Scholar - Xu HK: An iterative approach to quadratic optimization.
*J. Optim. Theory Appl.*2003, 116: 659–678. 10.1023/A:1023073621589View ArticleMathSciNetGoogle Scholar - Ambrosetti A, Prodi G:
*A Primer of Nonlinear Analysis*. Cambridge University Press, Cambridge; 1993.Google Scholar - Combettes PL, Hirstoaga SA: Equilibrium programming in Hilbert spaces.
*J. Nonlinear Convex Anal.*2005, 6: 117–136.MathSciNetGoogle Scholar - Reich S Theory and Applications of Nonlinear Operators of Accretive and Monotone Type. In
*A Weak Convergence Theorem for the Alternating Method with Bregman Distances*. Dekker, New York; 1996.Google Scholar - Borwein JM, Reich S, Sabach S: A characterization of Bregman firmly nonexpansive operators using a new monotonicity concept.
*J. Nonlinear Convex Anal.*2011, 12: 161–184.MathSciNetGoogle Scholar - Martín-Márquez V, Reich S, Sabach S: Right Bregman nonexpansive operators in Banach spaces.
*Nonlinear Anal.*2012, 75: 5448–5465. 10.1016/j.na.2012.04.048View ArticleMathSciNetGoogle Scholar - Martn-Mrquez V, Reich S, Sabach S: Bregman strongly nonexpansive operators in reflexive Banach spaces.
*J. Math. Anal. Appl.*2013, 400: 597–614. 10.1016/j.jmaa.2012.11.059View ArticleMathSciNetGoogle Scholar

## Copyright

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.