• No results found

An algorithm for a common fixed point of a family of pseudocontractive mappings

N/A
N/A
Protected

Academic year: 2020

Share "An algorithm for a common fixed point of a family of pseudocontractive mappings"

Copied!
14
0
0

Loading.... (view fulltext now)

Full text

(1)

R E S E A R C H

Open Access

An algorithm for a common fixed point of a

family of pseudocontractive mappings

Habtu Zegeye

1

and Naseer Shahzad

2*

*Correspondence:

[email protected]

2Department of Mathematics, King

Abdulaziz University, P.O. Box 80203, Jeddah, 21589, Saudi Arabia Full list of author information is available at the end of the article

Abstract

We introduce an iterative process which converges strongly to a common fixed point of a family of Lipschitzian pseudocontractive mappings. Our theorems improve and unify most of the results that have been proved for this important class of nonlinear operators.

MSC: 47H09; 47J20; 65J15

Keywords: fixed points of mappings; monotone mappings; pseudocontractive mappings; strong convergence

1 Introduction

LetCbe a nonempty subset of a real Hilbert space H. A mappingT :CH is called

pseudocontractiveif and only if

TxTy≤ xy+(IT)x– (IT)y, for allx,yC, (.)

andTis calledα-strictly pseudocontractive[] if and only if there existsα∈[, ) such that

Tx–Ty≤ x–y+α(IT)x– (IT)y, for allx,yC, (.)

Tis calledLipschitzianif and only if

Tx–Ty ≤Lxy, for allx,yK. (.)

If in (.) we haveL≤, thenTis callednonexpansive. We note that inequalities (.) and (.) can be equivalently written as

x–y,TxTy ≤ xy, for allx,yC (.)

and

xy,TxTyxy–λ(IT)x– (IT)y, (.)

for someλ> , respectively.

(2)

We observe from (.), (.) and (.) that every nonexpansive mapping is anα-strictly pseudocontractive mapping and everyα-strictly pseudocontractive mapping is a pseudo-contractive mapping, and hence a class of pseudopseudo-contractive mappings is a more general class of mappings.

Furthermore, pseudocontractive mappings are related with the important class of non-linearmonotonemappings, where a mappingAwith domainD(A) and rangeR(A) inHis calledmonotoneif the inequality

x–y,AxAy ≥, (.)

holds for everyx,yD(A). We note thatT is pseudocontractive if and only ifA:=IT

is monotone, and hence a fixed point of T, F(T) :={x∈D(T) :Tx=x}, is a zero ofA,

N(A) :={x∈D(A) :Ax= }. It is now well known (see,e.g., []) that ifAis monotone, then the solutions of the equationAx=  correspond to the equilibrium points of some evolution systems.

Finding a point in the intersection of fixed point sets of a family of nonexpansive map-pings is a task that occurs frequently in various areas of mathematical sciences and en-gineering. For example, the well-known convex feasibility problem reduces to finding a point in the intersection of fixed point sets of a family of nonexpansive mappings; see,e.g., [, ].

Consequently, considerable research efforts have been devoted to developing iterative methods for approximating a common fixed point (when it exists) for a family of nonex-pansive mappings andα-strictly pseudocontractive mappings. Bauschke [] was the first to introduce a Halpern-type iterative process (see,e.g., []) for approximating a common fixed point for a finite family ofNnonexpansive self-mappings. He proved the following theorem.

Theorem BSK (Bauschke [], Theorem .) Let K be a nonempty closed convex sub-set of a Hilbert space H, and let T,T, . . . ,TN be a finite family of nonexpansive

map-pings of K into itself with F:=Ni=F(Ti)=∅and F=F(TNTN–· · ·T) =F(TTN· · ·T) =

F(TN–TN–· · ·TTN).Given points u,x∈C,let{xn}be generated by

xn+:=αn+u+ ( –αn+)Tn+xn, n≥, (.)

where Tn:=Tn(modN)and{αn} ⊂(, )satisfies

n≥|αn+rαn|<∞.Then{xn}converges

strongly to PFu,where PF:HF is the metric projection.

Various authors have studied iterative schemes similar to that of Theorem BSK in more general Banach spaces on the one hand, and using various conditions on the sequence{αn} on the other hand (see, for example, Colaoet al.[], Yao [], Takahashi and Takahashi [], Plubtieng and Punpaeng [], Cenget al.[] and the references therein). Many authors have also studied iterative methods for a family ofα-strictly pseudocontractive mappings (see,e.g., [–] and the references therein).

Our concern now is the following:Can we construct an iterative sequence for a common fixed point of a finite family of pseudocontractive mappings?

(3)

Hilbert spaces. LetCbe a closed convex subset ofE, and let{Ti}Ni=be a finite family of Lipschitzian pseudocontractive self-mappings ofCsubset ofEwhich is a real uniformly convex Banach space with a Frêchet differentiable norm andF:=Ni=F(Ti)=∅. Let{xn} be defined by

xn=αnxn–+ ( –αn)Tnxn, n≥, (.)

whereTn=Tn(modN). He proved that{xn}converges weakly to a common fixed point of the family{Ti}Ni=under certain conditions on the parameter{αn}.

We remark that the scheme in the theorem of Zhou [] isimplicitand the convergence isweak convergence.

Recently, Zegeyeet al.[] proved the following strong convergence of Ishikawa iterative process [] for a common fixed point of a finite family of Lipschitz pseudocontractive mappings.

Theorem ZSA [] Let C be a nonempty, closed and convex subset of a real Hilbert space H.Let Ti:CC,i= , , . . . ,N,be a finite family of Lipschitz pseudocontractive

mappings with Lipschitzian constants Li,for i= , , . . . ,N,respectively.Assume that the interior of F:=Ni=F(Ti)is nonempty.Let{xn}be a sequence generated from an arbitrary

x∈E by ⎧ ⎨ ⎩

yn= ( –βn)xn+βnTnxn,

xn+= ( –αn)xn+αnTnyn, n≥,

(.)

where Tn:=Tn(modN) and{αn},{βn} ⊂(, )satisfy certain appropriate conditions.Then

{xn}converges strongly to a common fixed point of{T,T, . . . ,TN}.

It is worth to mention that the assumption ‘interior of F(T)is nonempty’ in Theo-rem ZSA issevere restriction.

More recently, Daman and Zegeye [] proved the following strong convergence of a Halpern-type [] iterative process for a common fixed point of a finite family of Lipschitz pseudocontractive mappings under the assumption that the family satisfiescondition(H), that is, the family{Ti:i= , , . . . ,N}of pseudocontractive mappings is said to satisfy

con-dition(H) if and only ifTixx,Tjxx ≥ fori,j∈ {, , . . . ,N}. We can also see that condition (H) is againsevere restriction.

It is our purpose in this paper to introduce an iterative scheme which converges strongly to a common fixed point of a family of Lipschitz pseudocontractive mappings. The as-sumption thatinterior of F(T)is nonemptyorcondition(H) is dispensed with. The results obtained in this paper improve and extend the results of Zhou [], Theorem ZSA, Daman and Zegeye [] and some other results in this direction.

2 Preliminaries

In what follows we shall make use of the following lemmas.

Lemma . Let H be a real Hilbert space.Then,for any given x,yE,the following in-equality holds:

(4)

Lemma .[] Let C be a convex subset of a real Hilbert space H.Let xH.Then x=

PCx if and only if

z–x,xx ≤, ∀z∈C.

Lemma .[] Let{an}be a sequence of nonnegative real numbers satisfying the following

relation:

an+≤( –αn)an+αnδn, nn,

where{αn} ⊂(, )and{δn} ⊂R satisfy the following conditions:limn→∞αn= ,

n=αn=

∞,andlim supn→∞δn≤.Thenlimn→∞an= .

Lemma .[] Let H be a real Hilbert space,C be a closed convex subset of H and T:

CC be a continuous pseudocontractive mapping.Then

(i) F(T)is a closed convex subset ofC;

(ii) (IT)is demiclosed at zero,i.e.,if{xn}is a sequence inCsuch thatxnxand

Txnxn→,asn→ ∞,thenx=T(x).

Lemma .[] Let{an}be sequences of real numbers such that there exists a subsequence

{ni}of{n}such that ani<ani+ for all iN.Then there exists a nondecreasing sequence

{mk} ⊂N such that mk→ ∞and the following properties are satisfied by all(sufficiently

large)numbers kN:

amkamk+ and akamk+.

In fact,mkis the largest number n in the set{, , . . . ,k}such that the condition anan+

holds.

Lemma .[] Let H be a real Hilbert space.Then,for all xiH andαi∈[, ]for

i= , , . . . ,n such thatα+α+· · ·+αn= ,the following equality holds:

αx+αx+· · ·+αnxn= n

i=

αixi–

≤i,jn

αiαjxixj.

3 Main result

Theorem . Let C be a nonempty,closed and convex subset of a real Hilbert space H.

Let T,T:CC be Lipschitz pseudocontractive mappings with Lipschitz constants L

and L,respectively.Assume thatF=F(T)∩F(T)is nonempty.Let a sequence{xn}be

generated from an arbitrary x,uC by

⎧ ⎪ ⎪ ⎨ ⎪ ⎪ ⎩

zn= ( –cn)xn+cnTxn;

yn= ( –βn)xn+βnTxn;

xn+=αnu+ ( –αn)(θnxn+δnTyn+γnTzn), n≥,

(.)

where{δn},{θn},{γn} ⊂(a,b)⊂(, ),{αn} ⊂(,c)⊂(, )satisfy the following conditions: (i)θn+δn+γn= ; (ii)limn→∞αn= ,

αn=∞; (iii)δnncn,βnβ<√ 

(5)

for L:=max{L,L}.Then{xn}converges strongly to a common fixed point of Tand T

nearest to u.

Proof We make use of some ideas of the paper []. LetpF. Then, from (.), (.) and

Lemma ., we have that

xn+–p =αnu+ ( –αn)(θnxn+δnTyn+γnTzn) –p

αnu–p+ ( –αn)δn(Tynp) +θn(xnp) +γn(Tznp) 

αnu–p+ ( –αn)

δnTynp+θnxnp

+γnTznp

– ( –αnnθnTynxn

– ( –αn)θnγnTznxn,

and hence

xn+–p≤αnu–p+ ( –αn)δn

ynp+ynTyn

+ ( –αn)θnxnp+ ( –αn)γn

znp+znTzn

– ( –αnnθnTynxn– ( –αnnγnTznxn

αnu–p+ ( –αn)δnynp+ ( –αn)δnynTyn

+ ( –αn)θnxnp+ ( –αn)γnznp

+ ( –αnnznTzn– ( –αnnθnTynxn

– ( –αnnγnTznxn. (.)

In addition, from (.), Lemma . and (.), we get that

ynp =( –βn)(xnp) +βn(Txnp) 

= ( –βn)xnp+βnTxnp

βn( –βn)xnTxn

≤( –βn)xnp+βn

xnp+xnTxn

βn( –βn)xnTxn

=xnp+βnxnTxn. (.)

Similarly, we have that

znp=xnp+cnxnTxn. (.)

Furthermore, from (.) and Lemma ., we have that

ynTyn=( –βn)(xnTyn) +βn(TxnTyn) 

(6)

βn( –βn)xnTxn

≤( –βn)xnTyn+βnLxnyn

βn( –βn)xnTxn

= ( –βn)xnTyn+βnLxnTxn

βn( –βn)xnTxn

= ( –βn)xnTyn–βn

 –Lβn–βnxnTxn. (.)

Similarly, we obtain that

znTzn= ( –cn)xnTzn–cn

 –Lcncn

xnTxn. (.)

Substituting (.), (.), (.) and (.) into (.), we obtain that

xn+–p≤αnu–p+ ( –αn)δn

xnp+βnxnTxn

+ ( –αn)δn

( –βn)xnTyn–βn

 –Lβn–βn

× xnTxn

+ ( –αnnxnp+ ( –αnn

xnp

+cnxnTxn

+ ( –αn)γn

( –cn)xnTzn

cn

 –Lcncn

xnTxn

– ( –αn)δnθnTynxn

– ( –αn)θnγnTznxn

=αnup+ ( –αn)xnp– ( –αnnβn

× –Lβn+ βn

xnTxn

– ( –αn)γncn

 –Lcn+ cn

xnTxn

+ ( –αn)δn( –θnβn)Tynxn

+ ( –αnn( –θncn)Tznxn,

and hence

xn+–p≤αnup+ ( –αn)xnp– ( –αnnβn

× –Lβn+ βn

xnTxn

– ( –αn)γncn

 –Lcn+ cn

xnTxn

+ ( –αn)δn(δn+γnβn)Tynxn

+ ( –αnnn+γncn)Tznxn. (.)

Now, from (iii) of the hypothesis, we have that

 – βn–Lβn≥ – β–Lβ> , (.)

(7)

and

n+γn) –βn≤, (δn+γn) –cn≤ for alln≥. (.)

Thus, inequality (.) implies that

xn+–p≤αnu–p+ ( –αn)xnp. (.)

Thus, by induction,

xn+–p≤max

up,x–p

, ∀n≥,

which implies that{xn}and hence{yn}are bounded.

Letx∗ =PF(u). Then, using (.), Lemma . and following the methods used to get (.), we obtain that

xn+–x∗ =αnu+ ( –αn)(θnxn+δnTyn+γnTzn) –x∗

n

ux∗+ ( –αn)θnxn+δnTyn+γnTznx∗ 

≤( –αn)δnTyn+θnxn+γnTznx∗ 

+ αn

ux∗,xn+–x

≤( –αnnTynx∗ 

+ ( –αnnxnx∗ 

+ ( –αn)γnTznx∗– ( –αn)θnδnTynxn

– ( –αn)θnγnTznxn+ αn

ux∗,xn+–x

and

xn+–x∗≤( –αn)δnynx∗+ynTyn

+ ( –αnnxnx∗ 

+ ( –αnnznx∗ 

+znTzn

– ( –αnnδnTynxn

– ( –αnnγnTznxn+ αn

ux∗,xn+–x

≤( –αn)δnxnx∗ 

+βnxnTxn

+ ( –αn)δn

( –βn)xnTyn–βn

 –Lβn–βn

× xnTxn

+ ( –αn)θnxnx∗ 

+ ( –αn)γnxnx∗ 

+cnxnTxn

+ ( –αn)γn

( –cn)xnTzn–cn

 –Lcncn

× xnTxn

– ( –αn)θnδnTynxn– ( –αn)θnγn

× Tznxn+ αn

ux∗,xn+–x

(8)

which implies that

xn+–x∗≤( –αn)xnx∗– ( –αn)δnβn

 –Lβn– βn

× xnTxn+ ( –αn)δn(δn+γnβn)xnTyn

– ( –αn)δncn

 –Lcn– cn

xnTxn

+ ( –αn)γn(δn+γncn)xnTzn+ αn

ux∗,xn+–x

(.)

≤( –αn)xnx∗+ αn

ux∗,xn+–x

. (.)

Now, we consider two cases.

Case . Suppose that there existsn∈Nsuch that{xnx∗}is decreasing for allnn. Then, we get that{xnx∗}is convergent. Thus, from (.), (.), (.) and (.), we have that

xnTxn→, xnTxn→ asn→ ∞. (.)

Furthermore, since{xn+} is a bounded subset ofH which is reflexive, we can choose a subsequence {xni+}of{xn+}such that xni+xandlim supn→∞u–x∗,xn+–x∗=

limi→∞ux∗,xni+–x∗. Then, from (.) and Lemma ., we have thatxF(T) and

xF(T). Therefore, by Lemma ., we immediately obtain that

lim sup n→∞

ux∗,xn+–x

= lim i→∞

ux∗,xni+–x

=ux∗,xx∗≤. (.)

Then it follows from (.), (.) and Lemma . thatxnx∗ → asn→ ∞. Conse-quently,xnx∗=PF(u).

Case . Suppose that there exists a subsequence{ni}of{n}such that

xnix∗<xni+–x

for alli∈N. Then, by Lemma ., there exists a nondecreasing sequence{mk} ⊂Nsuch thatmk→ ∞, and

xmkx∗≤xmk+–x∗ and xkx∗≤xmk+–x∗ (.)

for allk∈N. Now, from (.), (.), (.) and (.), we get thatxmkTxmk → and

xnkTxnk→ ask→ ∞. Thus, like in Case , we obtain that

lim sup k→∞

ux∗,xmk+–x

. (.)

Now, from (.) we have that

xmk+–x∗ 

≤( –αmk)xmkx∗ 

+ αmk

ux∗,xmk+–x

(9)

and hence (.) and (.) imply that

αmkxmkx∗ 

xmkx∗ 

xmk+–x∗ 

+ αmk

ux∗,xmk+–x

≤+αmk

ux∗,xmk+–x

.

But using the fact thatαmk>  and (.), we obtain that

lim sup k→∞

xmkx∗ 

≤,

and hencexmkx∗ → ask→ ∞. This together with (.) implies thatxmk+–x∗ →  ask→ ∞. Butxkx∗ ≤ xmk+–x

for allkN, thus we obtain thatx

kx∗. Therefore, from the above two cases, we can conclude that{xn}converges strongly to a common fixed point ofTandTnearest tou. The proof is complete.

We note that the method of the proof of Theorem . provides a convergence theorem for a finite family of Lipschitzian pseudocontractive mappings. In fact, we have the fol-lowing theorem.

Theorem . Let C be a nonempty,closed and convex subset of a real Hilbert space H.

Let Ti:CC,i= , , . . . ,N,be Lipschitz pseudocontractive mappings with Lipschitz

con-stants Li,i= , , . . . ,N,respectively.Assume thatF=Ni F(Ti)is nonempty.Let a sequence {xn}be generated from an arbitrary x,uC by

⎧ ⎨ ⎩

yni= ( –βn)xn+βnTixn, i= , , . . . ,N;

xn+=αnu+ ( –αn)(θnxn+ N

i=θniTiyni),

(.)

where{θni:i= , , , . . . ,N} ⊂(a,b)⊂(, ),{αn} ⊂(,c)⊂(, )satisfy the following

con-ditions: (i)θn+θn+· · ·+θnN= ; (ii)limn→∞αn= ,

αn=∞; (iii) N

i=θniβnβ< 

+L+,∀n≥,for L:=max{Li:i= , , . . . ,N}.Then{xn}converges strongly to a common fixed point of Ti(i= , , . . . ,N)nearest to u.

If in Theorem . we assume thatT is nonexpansive, then we have thatT is Lipschitz pseudocontractive withL= , and hence we get the following corollary.

Corollary . Let C be a nonempty,closed and convex subset of a real Hilbert space H.Let

T,T:CC be nonexpansive mappings.Assume thatF=F(T)∩F(T)is nonempty.

Let a sequence{xn}be generated from an arbitrary x,uC by

⎪ ⎪ ⎨ ⎪ ⎪ ⎩

zn= ( –cn)xn+cnTxn;

yn= ( –βn)xn+βnTxn;

xn+=αnu+ ( –αn)(θnxn+δnTyn+γnTzn),

(.)

where{δn},{θn},{γn} ⊂(a,b)⊂(, ),{αn} ⊂(,c)⊂(, )satisfy the following conditions: (i)δn+θn+γn= ; (ii)limn→∞αn= ,

αn=∞; (iii)δn+γncn,βnβ< (

 – ),∀n≥.

(10)

We now state and prove the convergence theorem for a common zero of a family of monotone mappings.

Corollary . Let H be a real Hilbert space.Let Ai:HH,i= , ,be Lipschitz monotone

mappings with Lipschitz constants Land L,respectively.Assume that F:= 

i=N(Ai)is

nonempty.Let a sequence{xn}be generated from an arbitrary x,uH by

⎪ ⎪ ⎨ ⎪ ⎪ ⎩

zn=xncnAxn;

yn=xnβnAxn;

xn+=αnu+ ( –αn)(θnxn+δn(IA)yn+γn(IA)zn),

(.)

where{δn},{θn},{γn} ⊂(a,b)⊂(, ),{αn} ⊂(,c)⊂(, )satisfy the following conditions: (i)δn+θn+γn= ; (ii)limn→∞αn= ,

αn=∞; (iii)δn+γncn,βnβ<√  +(+L)+, ∀n≥,for L:=max{L,L}.Then{xn}converges strongly to a common zero point of Aand

Anearest to u.

Proof LetTix:= (IAi)xfori= , . Then we get that everyTifor alli∈ {, }is a Lipschitz pseudocontractive mapping with the Lipschitz constant Li := ( +Li) and

i=F(Ti) = 

i=(Ai)=∅. Moreover, whenAiis replaced with (ITi), for eachi∈ {, }, then scheme (.) reduces to scheme (.), and hence the conclusion follows from Theorem ..

We may also have the following corollary for a finite family of monotone mappings.

Corollary . Let H be a real Hilbert space.Let Ai:HH,i= , , . . . ,N,be Lipschitz

monotone mappings with Lipschitz constants Li,i= , , . . . ,N,respectively.Assume that

F:=Ni=N(Ai)is nonempty.Let a sequence{xn}be generated from an arbitrary x,uH

by

⎧ ⎨ ⎩

yni=xnβnAixn, i= , , . . . ,N;

xn+=αnu+ ( –αn)(θnxn+ N

i=θni(IAi)yni),

(.)

where{θni:i= , , , . . . ,N} ⊂(a,b)⊂(, ),{αn} ⊂(,c)⊂(, )satisfy the following

con-ditions: (i)θn+θn+· · ·+θnN= ; (ii)limn→∞αn= ,

αn=∞; (iii) N

i=θniβnβ< 

+(+L)+,∀n≥,for L:=max{Li:i= , , . . . ,N}.Then{xn}converges strongly to a com-mon zero point of Ai,i= , , . . . ,N,nearest to u.

If in Corollary . we consider a single Lipschitz monotone mapping, then we obtain the following corollary.

Corollary . Let H be a real Hilbert space.Let A:HH be a Lipschitz monotone map-ping with Lipschitz constant L.Assume that F:=N(A)is nonempty.Let a sequence{xn}be

generated from an arbitrary x,uH by

⎧ ⎨ ⎩

yn=xnβnAxn;

xn+=αnu+ ( –αn)(( –γn)xn+γn(IA)yn),

(11)

where {γn} ⊂ (a,b) ⊂ (, ), {αn} ⊂ (,c) ⊂ (, ) satisfy the following conditions: (i)limn→∞αn= ,

αn=∞; (ii)γnβnβ< √ 

+(+L)+,∀n≥.Then{xn}converges strongly to the zero point of A nearest to u.

4 Numerical example

Now, we give an example of a finite family of pseudocontractive mappings satisfying The-orem . and some numerical experiment result to explain the conclusion of the theThe-orem as follows.

Example . LetH=Rwith absolute value norm. LetK= [–, ] andT,T:KKbe defined by

Tx:= ⎧ ⎨ ⎩

x+x, x[–, ],

x, x∈(, ] (.)

and

Tx:= ⎧ ⎨ ⎩

x, x∈[–,

],

x– (x–), x( , ].

(.)

Clearly,F=F(T)∩F(T) = [, ]∩[–,] = [,], and forx,yKwe have that

(IT)x– (IT)y,xy

≥ (.)

and

(IT)x– (IT)y,xy

≥, (.)

which show that both mappings are pseudocontractive. Next, we show that T is Lips-chitzian withL= . Ifx,y∈[–, ], then

|TxTy|=x+x–yy

=(x+y) + |x–y| ≤|x+y|. (.)

Ifx,y∈(, ], then

|TxTy|=|x–y|.

Ifx∈[–, ] andy∈(, ], then

|TxTy|=x+x–y

=xy+x=xy+x–y+y

=xy+x–y+y

≤ |x+y+ | · |x–y|+|y–x|

(12)

Ifx∈(, ] andy∈[–, ], then

|TxTy|=x

y+y

=xyy+x–x=xy+ (x+y)(xy) –x

=( +x+y)|x–y|+x

≤( +x+y)|x–y|+ (xy)

≤| +x+y|+|xy||xy|

≤|x–y|. (.)

Thus, we get thatTis Lipschitzian pseudocontractive withL= . Similarly, we can show thatTis Lipschitzian pseudocontractive withL= .

Now, takingαn=n+ ,cn=βn= +n+ ,γn=δn= .[ +n+ ] andθn=  – [ + 

n+], we observe that the conditions of Theorem . are satisfied and scheme (.) pro-vides the data in Tables  and  and Figures  and .

(i) Whenu= . andx= –, we see that the sequence converges tox∗= . as shown in Table  and Figure .

(ii) Whenu= – andx= ., we see that the sequence converges tox∗=  as shown in Table  and Figure .

Table 1 Values of{xn}with initial valuesu= 0.6 andx0= –1

n 0 500 1,000 5,000 10,000 12,000 14,000 17,000

xn –1.0000 0.5639 0.5638 0.5387 0.5287 0.5264 0.5246 0.5225

(13)

Table 2 Values of{xn}with initial valuesu= –1 andx0= 0.8

n 0 500 1,000 5,000 10,000 12,000 14,000 17,000

xn 0.8000 –0.4781 –0.3738 –0.1870 –0.1351 –0.1238 –0.1150 –0.1047

Figure 2 Figure of{xn}with initial valuesu= –1 andx0= 0.8.

Remark . Theorem . provides a convergence sequence to a common fixed point of two Lipschitzian pseudocontractive mappings, whereas Theorem . provides a conver-gence sequence to a common fixed point of a finite family of Lipschitzian pseudocontrac-tive mappings. In addition, Corollary . provides a convergence sequence to a common zero of two Lipschitzian monotone mappings, whereas Theorem . provides a conver-gence sequence to a common zero of a finite family of Lipschitzian monotone mappings.

Remark . Theorem . improves Theorem ZSA, Theorem . of Daman and Zegeye [] in the sense that our convergence does not require the assumption that interior of F(T)is nonemptyorcondition(H).

Remark . Theorem . improves Theorem . of Zhou [], Theorem . of Yaoet al.

[] and Theorem . of Tanget al.[] in the sense that our convergence requires neither compactness ofT nor computation of closed and convexCnofCfor eachn≥.

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

All the authors contributed equally. All authors read and approved the final manuscript.

Author details

1Department of Mathematics, University of Botswana, Pvt. Bag 00704, Gaborone, Botswana.2Department of

(14)

Acknowledgements

This article was funded by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah. The second author acknowledges with thanks DSR for financial support.

Received: 8 April 2013 Accepted: 15 August 2013 Published: 5 September 2013 References

1. Browder, FE, Petryshyn, WV: Construction of fixed points of nonlinear mappings in Hilbert space. J. Math. Anal. Appl. 20, 197-228 (1967)

2. Zeidler, E: Nonlinear Functional Analysis and Its Applications, Part II: Monotone Operators. Springer, Berlin (1985) 3. Bauschke, HH, Borwein, JM: On projection algorithms for solving convex feasibility problems. SIAM Rev.38, 367-426

(1996)

4. Combettes, PL: The foundations of set theoretic estimation. Proc. IEEE81, 182-208 (1993)

5. Bauschke, HH: The approximation of fixed points of compositions of nonexpansive mappings in Hilbert spaces. J. Math. Anal. Appl.202, 150-159 (1996)

6. Halpern, B: Fixed points of nonexpansive maps. Bull. Am. Math. Soc.73, 957-961 (1967)

7. Colao, V, Marino, G, Xu, HK: An iterative method for finding common solutions of equilibrium and fixed point problems. J. Math. Anal. Appl.344, 340-352 (2008)

8. Yao, Y: A general iterative method for a finite family of nonexpansive mappings. Nonlinear Anal.66, 2676-2687 (2007) 9. Takahashi, S, Takahashi, W: Viscosity approximation methods for equilibrium problems and fixed point problems in

Hilbert spaces. J. Math. Anal. Appl.331, 506-515 (2007)

10. Plubtieng, S, Punpaeng, R: A general iterative method for equilibrium problems and fixed point problems in Hilbert spaces. J. Math. Anal. Appl.336, 455-469 (2007)

11. Ceng, LC, Cubiotti, P, Yao, JC: Strong convergence theorems for finitely many nonexpansive mappings and applications. Nonlinear Anal.67, 1464-1473 (2007)

12. Osilike, MO, Shehu, Y: Cyclic algorithm for common fixed points of finite family of strictly pseudocontractive mappings of Browder-Petryshyn type. Nonlinear Anal.70, 3575-3583 (2009)

13. Zegeye, H, Shahzad, N: Strong convergence theorems for a common zero of a countably infinite family ofλ-inverse strongly accretive mappings. Nonlinear Anal.71, 531-538 (2009)

14. Zegeye, H, Shahzad, N: Strong convergence theorems for a common zero of a finite family ofm-accretive mappings. Nonlinear Anal.66, 1161-1169 (2007)

15. Zhang, Y, Guo, Y: Weak convergence theorems of three iterative methods for strictly pseudocontractive mappings of Browder-Petryshyn type. Fixed Point Theory Appl.2008, Article ID 672301 (2008)

16. Zhou, H: Convergence theorems of fixed points for a finite family of Lipschitz pseudo-contractions in Banach spaces. Nonlinear Anal.68, 2977-2983 (2008)

17. Zegeye, H, Shahzadb, N, Alghamdi, MA: Convergence of Ishikawa’s iteration method for pseudocontractive mappings. Nonlinear Anal.74, 7304-7311 (2011)

18. Ishikawa, S: Fixed points by a new iteration method. Proc. Am. Math. Soc.44, 147-150 (1974) 19. Daman, OA, Zegeye, H: Strong convergence theorems for a common fixed point of a finite family of

pseudocontractive mappings. Int. J. Math. Math. Sci.2012, Article ID 405315 (2012). doi:10.1155/2012/405315 20. Zhou, H: Convergence theorems of fixed points for Lipschitz pseudo-contractions in Hilbert spaces. J. Math. Anal.

Appl.343, 546-556 (2008)

21. Alber, Y: Metric and generalized projection operators in Banach spaces: properties and applications. In: Kartsatos, AG (ed.) Theory and Applications of Nonlinear Operators of Accretive and Monotone Type. Lecture Notes in Pure and Appl. Math., vol. 178, pp. 15-50. Dekker, New York (1996)

22. Xu, HK: Another control condition in an iterative method for nonexpansive mappings. Bull. Aust. Math. Soc.65, 109-113 (2002)

23. Zhang, QB, Cheng, CZ: Strong convergence theorem for a family of Lipschitz pseudocontractive mappings in a Hilbert space. Math. Comput. Model.48, 480-485 (2008)

24. Maingé, PE: Strong convergence of projected subgradient methods for nonsmooth and non-strictly convex minimization. Set-Valued Anal.16, 899-912 (2008)

25. Zegeye, H, Shahzad, N: Convergence of Mann’s type iteration method for generalized asymptotically nonexpansive mappings. Comput. Math. Appl.62, 4007-4014 (2011)

26. Yao, YH, Liou, YC, Marino, G: A hybrid algorithm for pseudo-contractive mappings. Nonlinear Anal.71, 4997-5002 (2009)

27. Tang, Y-C, Peng, J-G, Liu, L-W: Strong convergence theorem for pseudocontractive mappings in Hilbert spaces. Nonlinear Anal.74, 380-385 (2011). doi:10.1016/j.na.2010.08.048

doi:10.1186/1687-1812-2013-234

Figure

Table 1 Values of {xn} with initial values u = 0.6 and x0 = –1
Table 2 Values of {xn} with initial values u = –1 and x0 = 0.8

References

Related documents

Conclusions: The FINGER MRI exploratory sub-study did not show significant differences between the intervention and control groups on changes in regional brain volumes,

Four important issues emerge in fulfilling the demand being placed on pathology and radiology departments: (1) time spent at meetings, (2) timing and coordination of meetings, (3)

To investigate readthrough efficiency after intravenous gentamicin in CF patients with these mutations, in comparison with patients with other stop mutations and without stop

Results: Our analyses, based on the two commonly used mitochondrial genes cytochrome c oxidase I (the standard DNA barcode for animal species) and cytochrome b (a common

samples were incubated at 37 °C after spiking with A β 42 (10 ng/ml), and the presence of oligomers was measured using MDS at various time points after the start of incubation. As

In this large study based on the French National Alzheimer Database, the annual prevalence of AP prescribing among individuals with dementia slightly increased from 6.5% in 2010 to

The Hit rate here becomes a combination of the implicit action of the users (access to the cache without the knowledge that it contains the object) and the explicit one (access to