Skip to content

Advertisement

Open Access

On split common solution problems: new nonlinear feasible algorithms, strong convergence results and their applications

Fixed Point Theory and Applications20142014:219

https://doi.org/10.1186/1687-1812-2014-219

Received: 12 July 2014

Accepted: 26 September 2014

Published: 22 October 2014

Abstract

In this paper, we study and give examples for classes of generalized contractive mappings. We establish some new strong convergence theorems of feasible iterative algorithms for the split common solution problem (SCSP) and give some applications of these new results.

MSC:47H06, 47J25, 47H09, 65K10.

Keywords

Lipschitziandemicontractive mappingpseudocontractive mappingquasi-nonexpansive mappingnonexpansive mappingsplit common solution problem (SCSP)iterative algorithmstrong convergence theorem

1 Introduction and preliminaries

Let K be a closed convex subset of a real Hilbert space H with the inner product , and the norm . The following inequalities are known and useful.

  • x + y 2 y 2 + 2 x , x + y ;

  • x y 2 = x 2 + y 2 2 x , y for all x , y H ;

  • α x + ( 1 α ) y 2 = α x 2 + ( 1 α ) y 2 α ( 1 α ) x y 2 for all x , y H and α [ 0 , 1 ] .

For each point x H , there exists a unique nearest point in K, denoted by P K x , such that
x P K x x y for all  y K .
The mapping P K is called the metric projection from H onto K. It is well known that P K has the following properties:
  1. (i)

    x y , P K x P K y P K x P K y 2 for every x , y H .

     
  2. (ii)

    For x H and z K , z = P K x x z , z y 0 for all y K .

     
  3. (iii)
    For x H and y K ,
    y P K x 2 + x P K x 2 x y 2 .
    (1.1)
     
Let H 1 and H 2 be two Hilbert spaces. Let A : H 1 H 2 and A : H 2 H 1 be two bounded linear operators. A is called the adjoint operator (or adjoint) of A if
A z , w = z , A w for all  z H 1  and  w H 2 .

It is known that the adjoint operator of a bounded linear operator on a Hilbert space always exists and is bounded linear and unique. Moreover, it is not hard to show that if A is an adjoint operator of A, then A = A . The symbols and are used to denote the sets of positive integers and real numbers, respectively.

Let H 1 and H 2 be two real Hilbert spaces. Let C be a closed convex subset of H 1 and K be a closed convex subset of H 2 . Let T : C C with F ( T ) and S : K K with F ( S ) be two mappings. Let A : H 1 H 2 be a bounded linear operator. The mathematical model of the split common solution problem (SCSP in short) is defined as follows:
( SCSP ) Find  p C  such that  T p = p  and  u : = A p K  satisfying  S u = u .

In fact, SCSP contains several important problems as special cases and many authors have studied and introduced some new iterative algorithms for SCSP and presented some strong and weak convergence theorems for SCSP; see, for instance, [124] and the references therein. Motivated and inspired by their works, in this paper, we study and establish new strong convergence results by using new iterative algorithms of SCSP for pseudocontractive mappings and k-demicontractive mappings in Hilbert spaces.

The paper is divided into four sections. In Section 2, we study and give examples for classes of generalized contractive mappings. Some new strong convergence theorems of feasible iterative algorithms for SCSP are established in Section 3. Finally, some applications and further remarks for our new results are given in Section 4. Consequently, in this paper, some of our results are original and completely different from these known related results in the literature.

2 Classes of generalized contractive mappings and their examples

Let T be a mapping with domain D ( T ) and range R ( T ) in a Hilbert space H. Recall that T is said to be
  1. (i)
    pseudocontractive if
    T x T y , x y x y 2 , x , y D ( T ) ,
    or, equivalently,
    T x T y 2 x y 2 + ( I T ) x ( I T ) y 2 , x , y D ( T ) ;
     
  2. (ii)
    demicontractive if, for all x D ( T ) and p F ( T ) ,
    T x p , x p x p 2
     
or, equivalently,
T x p 2 x p 2 + ( I T ) x 2 ;
  1. (iii)
    k-demicontractive if there exists a constant k [ 0 , 1 ) such that
    T x p 2 x p 2 + k ( I T ) x 2 for all  x D ( T )  and  p F ( T ) ;
     
  2. (iv)
    quasi-nonexpansive if it is 0-demicontractive, that is,
    T x p x p for all  x D ( T )  and  p F ( T ) ;
     
  3. (v)
    Lipschitzian if there exists L > 0 such that
    T x T y L x y , x , y D ( T ) ;
     
  4. (vi)

    nonexpansive if it is Lipschitzian with L = 1 ;

     
  5. (vii)

    contractive if it is Lipschitzian with L < 1 .

     
A Banach space ( X , ) is said to satisfy Opial’s condition if, for each sequence { x n } in X which converges weakly to a point x X , we have
lim inf n x n x < lim inf n x n y , y X , y x .

It is well known that any Hilbert space satisfies Opial’s condition.

Definition 2.1 (see [2])

Let K be a nonempty closed convex subset of a real Hilbert space H and T be a mapping from K into K. The mapping T is said to be demiclosed if, for any sequence { x n } which weakly converges to y, and if the sequence { T x n } strongly converges to z, then T y = z .

Remark 2.1 In Definition 2.1, the particular case of demiclosedness at zero is frequently used in some iterative convergence algorithms, which is the particular case when z = θ , the zero vector of H; for more details, one can refer to [2].

The following concept of zero-demiclosedness was introduced as follows.

Definition 2.2 (see [[25], Definition 2.3])

Let K be a nonempty closed convex subset of a real Hilbert space and T be a mapping from K into K. The mapping T is called zero-demiclosed if { x n } in K satisfying x n T x n 0 and x n z K implies T z = z .

The following result was essentially proved in [25], but we give the proof for the sake of completeness and the reader’s convenience.

Theorem 2.1 (see [[25], Proposition 2.4])

Let K be a nonempty closed convex subset of a real Hilbert space with zero vector θ. Then the following statements hold.
  1. (a)

    Let T be a mapping from K into K. Then T is zero-demiclosed if and only if I T is demiclosed at θ.

     
  2. (b)

    Let T be a nonexpansive mapping from H into itself. If there is a bounded sequence { x n } H such that x n T x n 0 as n 0 , then T is zero-demiclosed.

     
Proof Obviously, the conclusion (a) holds. To see (b), since { x n } is bounded, there is a subsequence { x n k } { x n } and z H such that x n k z . One can claim T z = z . Indeed, if T z z , it follows from Opial’s condition that
lim inf k x n k z < lim inf k x n k T z lim inf k { x n k T x n k + T x n k T z } = lim inf k T x n k T z lim inf k x n k z ,

which is a contradiction. So T z = z and hence T is zero-demiclosed. □

Now, we give some examples to show the existence of these generalized contractive mappings (i)-(vi) which also expound the relation between them.

Example A Let H = R with the absolute-value norm | | and C = [ 2 , 0 ] . Let T : C C be defined by
T x = { x 2 2 if  x [ 1 , 0 ] , 1 if  x [ 2 , 1 ] .
Then F ( T ) = { 1 } . Since
| T x ( 1 ) | 2 | x ( 1 ) | 2 + 1 2 | T x x | 2 for all  x C ,
we know that T is a 1 2 -demicontractive mapping. However, due to
| T ( 1 2 ) ( 1 ) | > | 1 2 ( 1 ) | ,

T is not quasi-nonexpansive.

Example B Let H = R with the absolute-value norm | | and C = [ 1 2 , 2 ] . Let T : C C be defined by
T x = 1 x , x C .
Then F ( T ) = { 1 } . Since
| T x 1 | 2 | x 1 | 2 + 3 4 | T x x | 2 for all  x C ,

T is a 3 4 -demicontractive mapping. Moreover, T is also a pseudocontractive mapping.

Example C Let H = R with the absolute-value norm | | . Let T : H H be defined by
T x = { ( 1 + x ) if  x 2 , x + 1 if  x 2 .
It is easy to see that
| T x T y | | x y | for all  x , y H .

So T is continuous nonexpansive with F ( T ) = .

The following example shows that there exists a continuous quasi-nonexpansive mapping which is not nonexpansive.

Example D (see [8])

Let H = R with the absolute-value norm | | and C = [ 0 , + ) . Define T : C C by
T x = x 2 + 2 1 + x , x C .
Obviously, F ( T ) = { 2 } . It is easy to see that
| T x 2 | = x 1 + x | x 2 | | x 2 | for all  x C
and
| T ( 0 ) T ( 1 3 ) | = 5 12 > | 0 1 3 | .

Hence T is a continuous quasi-nonexpansive mapping but not nonexpansive.

The following example shows that there exists a demicontractive mapping which is neither pseudocontractive nor k-demicontractive for all k [ 0 , 1 ) .

Example E Let H = R with the absolute-value norm | | . Let T : H H be defined by
T x = { x 2 x + 1 if  x ( , 1 ] , x 2 + 1 1 + x if  x [ 1 , + ) .
Then F ( T ) = { 1 } . Since
| T x 1 | 2 | x 1 | 2 + | T x x | 2 for all  x H ,
T is a demicontractive mapping. However, T is not a pseudocontractive mapping due to the fact that when x = 3 and y = 2.5 , we have
| T x T y | 2 > | x y | 2 + | ( x T x ) ( y T y ) | 2 .

It is easy to see that T is not a k-demicontractive mapping for all k [ 0 , 1 ) .

The following example shows that there exists a discontinuous pseudocontractive mapping which is not a demicontractive mapping.

Example F Let H = R with the absolute-value norm | | . Let T : H H be defined by
T x = { x 2 + 1 if  x ( , 0 ] , 1 x 2 if  x ( 0 , + ) .
Then F ( T ) = . Due to
| T x T y | 2 | x y | 2 + | ( I T ) x ( I T ) y | 2 for all  x H ,

we know that T is a discontinuous pseudocontractive mapping but not a demicontractive mapping.

The following example shows that there exists a pseudocontractive mapping which is not k-demicontractive for all k [ 0 , 1 ) .

Example G Let H = R with the absolute-value norm | | . Let T : H H be defined by
T x = { 2 x 2 if  x [ 0 , 1 ] , 2 x if  x [ 1 , 2 ] , 0 if  x [ 1 , + ) .
Then F ( T ) = { 1 } . Since
| T x T y | 2 | x y | 2 + | ( I T ) x ( I T ) y | 2 for all  x H ,

T is a pseudocontractive mapping. It is easy to see that T is not a k-demicontractive mapping for all k [ 0 , 1 ) .

The following example shows that there exists a discontinuous k-demicontractive mapping for some k [ 0 , 1 ) as well as being demiclosed at θ which is neither pseudocontractive nor quasi-nonexpansive.

Example H Let H = R with the absolute-value norm | | and C = [ 2 , 0 ] . Let T : C C be defined by
T x = { x 2 2 if  x [ 1 , 0 ] , 1 8 if  x = 3 2 , 1 if  x [ 2 , 3 2 ) ( 3 2 , 1 ] .
Then the following statements hold.
  1. (a)

    T is discontinuous 3 4 -demicontractive.

     
  2. (b)

    T is demiclosed at θ.

     
  3. (c)

    T is not pseudocontractive.

     
  4. (d)

    T is not quasi-nonexpansive.

     
Proof Clearly, F ( T ) = { 1 } . Since
| T x ( 1 ) | 2 | x ( 1 ) | 2 + 3 4 | ( I T ) x | 2 for all  x C ,
T is a discontinuous 3 4 -demicontractive mapping and (a) is proved. Now, we verify (b). In fact, let { x n } [ 2 , 0 ] with x n z and x n T x n 0 as n . If all x n [ 1 , 0 ] , we can prove T z = z and z = 1 F ( T ) easily. If there exists a subsequence { x n k } [ 2 , 1 ] , then, from x n T x n 0 as n , we can find a subsequence { x n k i } of { x n k } such that x n k i 3 2 for all i. Hence we have
| z ( 1 ) | | z x n k i | + | x n k i T x n k i | + | T x n k i ( 1 ) | 0 as  i ,
which implies z = 1 F ( T ) . To see (c) and (d), note that
| T ( 3 2 ) T ( 25 16 ) | 2 > | 3 2 ( 25 16 ) | 2 + | ( I T ) ( 3 2 ) ( I T ) ( 25 16 ) | 2
and
| T ( 3 2 ) ( 1 ) | > | ( 3 2 ) ( 1 ) | ,

so T is neither pseudocontractive nor quasi-nonexpansive. The proof is completed. □

3 New feasible iterative algorithms for SCSP and strong convergence theorems

In this section, we establish some new strong convergence theorems by using feasible iterative algorithms for SCSP.

Theorem 3.1 Let H 1 and H 2 be two real Hilbert spaces and θ i be the zero vector of H i for i = 1 , 2 . Let C be a nonempty closed convex subset of H 1 and A : H 1 H 2 be a bounded linear operator with its adjoint B. Let T : C C be a Lipschitzian pseudocontractive mapping with Lipschitz constant L > 0 and F ( T ) , and let S : H 2 H 2 be a k-demicontractive mapping with F ( S ) which is demiclosed at θ 2 . Let C 1 = C and { x n } be a sequence generated by the following algorithm:
{ x 1 C 1 chosen arbitrarily , y n = ( 1 α ) x n + α T x n , z n = β x n + ( 1 β ) T y n , w n = P C ( z n + ξ B ( S I ) A z n ) , C n + 1 = { v C n : w n v z n v x n v } , x n + 1 = P C n + 1 ( x 1 ) , n N ,
(3.1)
where 0 < 1 β < α < 1 2 1 + L 2 , ξ ( 0 , 1 k B 2 ) and P C n is the projection operator from H 1 into C n for n N . Suppose that
Ω = { p F ( T ) : A p F ( S ) } .
Then there exists q Ω such that
  1. (a)

    x n q as n ,

     
  2. (b)

    A x n A q as n .

     

Proof We will show the conclusion by proceeding with the following steps.

Step 1. For any p Ω , we prove
w n p 2 z n p 2 ξ ( 1 k ξ B 2 ) ( S I ) A z n 2 .
(3.2)
Indeed, since
w n p 2 z n + ξ B ( S I ) A z n p 2 = z n p 2 + ξ B ( S I ) A z n 2 + 2 ξ z n p , B ( S I ) A z n = z n p 2 + ξ B ( S I ) A z n 2 + 2 ξ A z n A p , ( S I ) A z n = z n p 2 + ξ B ( S I ) A z n 2 + 2 ξ A z n A p + ( S I ) A z n ( S I ) A z n , ( S I ) A z n = z n p 2 + ξ B ( S I ) A z n 2 + 2 ξ S A z n A p , ( S I ) A z n 2 ξ ( S I ) A z n 2 z n p 2 + ξ 2 B 2 ( S I ) A z n 2 + 2 ξ S A z n A p , ( S I ) A z n 2 ξ ( S I ) A z n 2
and
2 ξ S A z n A p , ( S I ) A z n = ξ { S A z n A p 2 + ( S I ) A z n 2 A z n A p 2 } ξ { A z n A p 2 + k ( S I ) A z n 2 + ( S I ) A z n 2 A z n A p 2 } ξ { A z n A z n 2 + k ( S I ) A z n 2 + ( S I ) A z n 2 } = ξ { k ( S I ) A z n 2 + ( S I ) A z n 2 } ,
we get
w n p 2 z n p 2 ξ ( 1 k ξ B 2 ) ( S I ) A z n 2 ,

and our desired result is proved.

Step 2. We prove
z n p x n p for all  n N .
(3.3)
For any n N , by (3.1), we have
z n p 2 = β x n p 2 + ( 1 β ) T y n p 2 ( 1 β ) β T y n x n 2 β x n p 2 + ( 1 β ) y n p 2 + ( 1 β ) T y n y n 2 ( 1 β ) β T y n x n 2 β x n p 2 + ( 1 β ) ( ( 1 α ) x n p 2 + α T x n x n 2 ( 1 α ) α T x n x n 2 ) + ( 1 β ) T y n y n 2 ( 1 β ) β T y n x n 2 x n p 2 + ( 1 β ) ( α T x n x n 2 ( 1 α ) α T x n x n 2 ) ( 1 β ) β T y n x n 2 + ( 1 β ) ( 1 α ) ( x n T y n ) + α ( T x n T y n ) 2 x n p 2 + ( 1 β ) ( α T x n x n 2 ( 1 α ) α T x n x n 2 ) ( 1 β ) β T y n x n 2 + ( 1 β ) ( ( 1 α ) x n T y n 2 + α T x n T y n 2 ( 1 α ) α T x n x n 2 ) x n p 2 + ( 1 β ) ( α T x n x n 2 ( 1 α ) α T x n x n 2 ) ( 1 β ) β T y n x n 2 + ( 1 β ) ( ( 1 α ) x n T y n 2 + α L 2 x n y n 2 ( 1 α ) α T x n x n 2 ) x n p 2 + ( 1 β ) ( α T x n x n 2 ( 1 α ) α T x n x n 2 ) ( 1 β ) β T y n x n 2 + ( 1 β ) ( ( 1 α ) x n T y n 2 + α 3 L 2 x n T x n 2 ( 1 α ) α T x n x n 2 ) = x n p 2 ( 1 β ) ( α + β 1 ) T y n x n 2 ( 1 β ) α ( 1 2 α α 2 L 2 ) T x n x n 2 .
(3.4)
Since α + β > 1 and α < 1 2 1 + L 2 , from (3.4), we have z n p 2 x n p 2 , or, equivalently,
z n p x n p .
(3.5)

Step 3. We show that C n is a nonempty closed convex set for any n N .

For any p Ω , by taking into account (3.2) and (3.5), we obtain
w n p z n p x n p for all  n N .

So we know Ω C n and hence C n for all n N . It is easy to verify that C n is closed and convex for all n N .

Step 4. We prove that { x n } is a Cauchy sequence in C and x n q as n for some q C .

Since Ω C n + 1 C n and x n + 1 = P C n + 1 ( x 1 ) C n , we get
x n + 1 x 1 p x 1 for all  p Ω
and
x n x 1 x n + 1 x 1 for all  n N ,
which show that { x n } is bounded and { x n x 1 } is nondecreasing in [ 0 , ) . So
lim n x n x 1 0
exists. For any m , n N with m > n , from x m = P C m ( x 1 ) C n and (1.1), we have
x m x n 2 + x 1 x n 2 = x m P C n ( x 1 ) 2 + x 1 P C n ( x 1 ) 2 x m x 1 2 .
(3.6)
Inequality (3.6) implies
lim m , n x n x m = 0 .
So { x n } is a Cauchy sequence. Clearly,
lim n x n + 1 x n = 0 .
(3.7)

By the completeness of C, there exists q C such that x n q as n .

Step 5. Finally, we show that the following hold:
  1. (i)

    q Ω ,

     
  2. (ii)

    A x n A q as n .

     
For any n N , since x n + 1 = P C n + 1 ( x 1 ) C n + 1 C n , from (3.1), we have
z n x n z n x n + 1 + x n + 1 x n 2 x n + 1 x n
(3.8)
and
w n x n w n x n + 1 + x n + 1 x n 2 x n + 1 x n .
(3.9)
From inequalities (3.7), (3.8) and (3.9), we deduce
lim n z n x n = 0 , lim n w n x n = 0
(3.10)
and hence
lim n w n z n = 0 .
(3.11)
By taking into account (3.4) and (3.10), we get
α ( 1 2 α α 2 L 2 ) T x n x n 2 + ( α + β 1 ) T y n x n 2 1 1 β ( x n p 2 z n p 2 ) 2 1 β x n z n x n p 0 as  n .
So, we obtain
lim n T x n x n = lim n T y n x n = 0 .
(3.12)
Since x n q as n , from (3.12) and the continuity of the norm and the Lipschitzian pseudocontractive mapping T, we can deduce that T q = q , namely q F ( T ) . On the other hand, from (3.2) and (3.11), we have
ξ ( 1 k ξ B 2 ) ( S I ) A z n 2 z n p 2 w n p 2 z n w n ( z n p w n p ) 0 as  n ,
which yields that
lim n ( S I ) A z n = 0 .
(3.13)
Since the k-demicontractive mapping S is demiclosed at θ 2 , taking into account x n q , A x n A q , z n x n 0 and (3.13), we have
A z n A q
and
A q F ( S ) .

Hence we confirm q Ω . The proof is completed. □

By virtue of Theorem 3.1, we can establish the following:
  1. (i)

    Strong convergence algorithms for the split common solution problem for Lipschitzian pseudocontractive mappings and nonexpansive mappings (see Corollary 3.1 below).

     
  2. (ii)

    Strong convergence algorithms for the split common solution problem for Lipschitzian pseudocontractive mappings and quasi-nonexpansive mappings (see Corollary 3.2 below).

     
Corollary 3.1 Let H 1 and H 2 be two real Hilbert spaces and θ i be the zero vector of H i for i = 1 , 2 . Let C be a nonempty closed convex subset of H 1 and A : H 1 H 2 be a bounded linear operator with its adjoint B. Let T : C C be a Lipschitzian pseudocontractive mapping with Lipschitz constant L > 0 and F ( T ) , and let S : H 2 H 2 be a nonexpansive mapping with F ( S ) . Let C 1 = C and { x n } be a sequence generated by the following algorithm:
{ x 1 C 1 chosen arbitrarily , y n = ( 1 α ) x n + α T x n , z n = β x n + ( 1 β ) T y n , w n = P C ( z n + ξ B ( S I ) A z n ) , C n + 1 = { v C n : w n v z n v x n v } , x n + 1 = P C n + 1 ( x 1 ) , n N ,
where 0 < 1 β < α < 1 2 1 + L 2 , ξ ( 0 , 1 B 2 ) and P C n is the projection operator from H 1 into C n for n N . Suppose that
Ω = { p F ( T ) : A p F ( S ) } .
Then there exists q Ω such that
  1. (a)

    x n q as n ,

     
  2. (b)

    A x n A q as n .

     

Proof Since the mapping S is nonexpansive, it is 0-demicontractive. Hence the desired conclusion follows from Theorem 3.1 immediately by taking k = 0 . □

Corollary 3.2 Let H 1 and H 2 be two real Hilbert spaces and θ i be the zero vector of H i for i = 1 , 2 . Let C be a nonempty closed convex subset of H 1 and A : H 1 H 2 be a bounded linear operator with its adjoint B. Let T : C C be a Lipschitzian pseudocontractive mapping with Lipschitz constant L > 0 and F ( T ) , and let S : H 2 H 2 be a quasi-nonexpansive mapping with F ( S ) which is demiclosed at θ 2 . Let C 1 = C and { x n } be a sequence generated by the following algorithm:
{ x 1 C 1 chosen arbitrarily , y n = ( 1 α ) x n + α T x n , z n = β x n + ( 1 β ) T y n , w n = P C ( z n + ξ B ( S I ) A z n ) , C n + 1 = { v C n : w n v z n v x n v } , x n + 1 = P C n + 1 ( x 1 ) , n N ,
where 0 < 1 β < α < 1 2 1 + L 2 , ξ ( 0 , 1 B 2 ) and P C n is the projection operator from H 1 into C n for n N . Suppose that
Ω = { p F ( T ) : A p F ( S ) } .
Then there exists q Ω such that
  1. (a)

    x n q as n ,

     
  2. (b)

    A x n A q as n .

     
Example 3.1 Let H 1 = R with the absolute-value norm | | . Let H 2 = [ 1 2 , 2 ] 2 with the norm α = ( a 1 2 + a 2 2 ) 1 2 for α = ( a 1 , a 2 ) H 2 and the inner product α , β = i = 1 2 a i b i for α = ( a 1 , a 2 ) and β = ( b 1 , b 2 ) H 2 . Let A : H 1 H 2 be defined by A x = ( x , x ) for x R . Then A is a bounded linear operator with its adjoint operator B z = z 1 + z 2 for z = ( z 1 , z 2 ) H 2 . Clearly, A = B = 2 . Let C = [ 1 2 , 2 ] . Let T : C C and S : H 2 H 2 be defined by
T x = 1 x for  x C
and
S z = ( 1 z 1 , 1 z 2 ) for  z = ( z 1 , z 2 ) H 2 ,

respectively. It is easy to see that

  • F ( T ) = { 1 } ;

  • F ( S ) = { ( 1 , 1 ) } ;

  • Ω = { p F ( T ) : A p F ( S ) } = { 1 } ;

  • T is a Lipschitzian pseudocontractive mapping with Lipschitz constant L = 2 ;

  • T and S both are 3 4 -demicontractive mappings.

By using algorithm (3.1) with 0 < 1 β < α < 1 2 3 and ξ ( 0 , 1 8 ) , we can verify x n 1 and A x n A ( 1 ) = ( 1 , 1 ) F ( S ) as n .

4 Some applications and further remarks for Theorem 3.1

Let C be a nonempty subset of a Hilbert space H. Recall that a mapping U : C C is said to be accretive if
U x U y , x y 0 for all  x , y C .
Obviously, U : C C is accretive if and only if I U : C C is pseudocontractive. Moreover,
F ( I U ) = U 1 ( θ ) : = { x C : U x = θ } ,

where θ is the zero vector of H.

At the end of this paper, by applying Theorem 3.1, we obtain the following:
  1. (i)

    Strong convergence algorithms for the split common solution problem for Lipschitzian accretive mappings and demicontractive nonexpansive mappings (see Theorem 4.1 below).

     
  2. (ii)

    Strong convergence algorithms for the split common solution problem for Lipschitzian accretive mappings and nonexpansive mappings (see Corollary 4.1 below).

     
  3. (iii)

    Strong convergence algorithms for the split common solution problem for Lipschitzian accretive mappings and quasi-nonexpansive mappings (see Corollary 4.2 below).

     
Theorem 4.1 Let H 1 and H 2 be two real Hilbert spaces and θ i be the zero vector of H i for i = 1 , 2 . Let A : H 1 H 2 be a bounded linear operator with its adjoint B and U : H 1 H 1 be a Lipschitzian accretive mapping with Lipschitz constant L > 0 and U 1 ( θ 1 ) . Let S : H 2 H 2 be a k-demicontractive mapping with F ( S ) which is demiclosed at θ 2 . Let { x n } be a sequence generated by the following algorithm:
{ x 1 H 1 chosen arbitrarily , y n = x n α U x n , z n = β x n + ( 1 β ) ( I U ) y n , w n = z n + ξ B ( S I ) A z n , C n + 1 = { v C n : w n v z n v x n v } , x n + 1 = P C n + 1 ( x 1 ) , n N ,
(4.1)
where 0 < 1 β < α < 1 2 1 + L 2 , ξ ( 0 , 1 k B 2 ) and P C n is the projection operator from H 1 into C n for n N . Suppose that
Ω = { p U 1 ( θ 1 ) : A p F ( S ) } .
Then there exists q Ω such that
  1. (a)

    x n q as n ,

     
  2. (b)

    A x n A q as n .

     
Proof Let C 1 = H 1 . Then the iterative process (4.1) can be rewritten as follows:
{ x 1 C 1 chosen arbitrarily , y n = ( 1 α ) x n + α ( I U ) x n , z n = β x n + ( 1 β ) ( I U ) y n , w n = z n + ξ B ( S I ) A z n , C n + 1 = { v C n : w n v z n v x n v } , x n + 1 = P C n + 1 ( x 1 ) , n N .

Set T : = I U , then F ( T ) = U 1 ( θ 1 ) and T is a Lipschitzian pseudocontractive mapping with Lipschitz constant 1 + L . Therefore the desired conclusion follows from Theorem 3.1 immediately. □

The following interesting results are immediate from Theorem 4.1.

Corollary 4.1 Let H 1 and H 2 be two real Hilbert spaces and θ i be the zero vector of H i for i = 1 , 2 . Let A : H 1 H 2 be a bounded linear operator with its adjoint B and U : H 1 H 1 be a Lipschitzian accretive mapping with Lipschitz constant L > 0 and U 1 ( θ 1 ) . Let S : H 2 H 2 be a quasi-nonexpansive mapping with F ( S ) which is demiclosed at θ 2 . Let { x n } be a sequence generated by the following algorithm:
{ x 1 H 1 chosen arbitrarily , y n = x n α U x n , z n = β x n + ( 1 β ) ( I U ) y n , w n = z n + ξ B ( S I ) A z n , C n + 1 = { v C n : w n v z n v x n v } , x n + 1 = P C n + 1 ( x 1 ) , n N ,
where 0 < 1 β < α < 1 2 1 + L 2 , ξ ( 0 , 1 B 2 ) and P C n is the projection operator from H 1 into C n for n N . Suppose that
Ω = { p U 1 ( θ 1 ) : A p F ( S ) } .
Then there exists q Ω such that
  1. (a)

    x n q as n ,

     
  2. (b)

    A x n A q as n .

     
Corollary 4.2 Let H 1 and H 2 be two real Hilbert spaces and θ i be the zero vector of H i for i = 1 , 2 . Let A : H 1 H 2 be a bounded linear operator with its adjoint B. Let U : H 1 H 1 be a Lipschitzian accretive mapping with Lipschitz constant L > 0 and U 1 ( θ 1 ) . Let S : H 2 H 2 be a nonexpansive mapping with F ( S ) . Let { x n } be a sequence generated by the following algorithm:
{ x 1 H 1 chosen arbitrarily , y n = x n α U x n , z n = β x n + ( 1 β ) ( I U ) y n , w n = z n + ξ B ( S I ) A z n , C n + 1 = { v C n : w n v z n v x n v } , x n + 1 = P C n + 1 ( x 1 ) , n N ,
where 0 < 1 β < α < 1 2 1 + L 2 , ξ ( 0 , 1 B 2 ) and P C n is the projection operator from H 1 into C n for n N . Suppose that
Ω = { p U 1 ( θ 1 ) : A p F ( S ) } .
Then there exists q Ω such that
  1. (a)

    x n q as n ,

     
  2. (b)

    A x n A q as n .

     

Remark 4.1 In Theorems 3.1 and 4.1, the control coefficients α and β can be respectively replaced with the sequences { α n } and { β n } satisfying 0 < ε < 1 β n < α n < 1 2 1 + L 2 for some positive real number ε.

Remark 4.2 Obviously, all results in this paper are true if H 1 = H 2 . They generalize and improve many results in the literature; see, for instance, [23, 24, 2629].

Declarations

Acknowledgments

The first author was supported by the Candidate Foundation of Youth Academic Experts at Honghe University (2014HB0206); the second author was supported by Grant No. MOST 103-2115-M-017-001 of the Ministry of Science and Technology of the Republic of China.

Authors’ Affiliations

(1)
Department of Mathematics, Honghe University, Mengzi, China
(2)
Department of Mathematics, National Kaohsiung Normal University, Kaohsiung, Taiwan

References

  1. Censor Y, Segal A: The split common fixed point problem for directed operators. J. Convex Anal. 2009, 16: 587–600.MathSciNetGoogle Scholar
  2. Moudafi A: A note on the split common fixed-point problem for quasi-nonexpansive operators. Nonlinear Anal. 2011, 74: 4083–4087. 10.1016/j.na.2011.03.041View ArticleMathSciNetGoogle Scholar
  3. Moudafi A: Split monotone variational inclusions. J. Optim. Theory Appl. 2011, 150: 275–283. 10.1007/s10957-011-9814-6View ArticleMathSciNetGoogle Scholar
  4. Yao Y, Postolache M, Liou Y-C: Strong convergence of a self-adaptive method for the split feasibility problem. Fixed Point Theory Appl. 2013., 2013: Article ID 201Google Scholar
  5. Censor Y, Gibali A, Reich S: Algorithms for the split variational inequality problem. Numer. Algorithms 2012, 59(2):301–323. 10.1007/s11075-011-9490-5View ArticleMathSciNetGoogle Scholar
  6. Zhao J, He S: Strong convergence of the viscosity approximation process for the split common fixed-point problem of quasi-nonexpansive mapping. J. Appl. Math. 2012., 2012: Article ID 438023 10.1155/2012/438023Google Scholar
  7. He Z: The split equilibrium problems and its convergence algorithms. J. Inequal. Appl. 2012., 2012: Article ID 162Google Scholar
  8. He Z, Du W-S: Nonlinear algorithms approach to split common solution problems. Fixed Point Theory Appl. 2012., 2012: Article ID 130Google Scholar
  9. He Z, Du W-S: On hybrid split problem and its nonlinear algorithms. Fixed Point Theory Appl. 2013., 2013: Article ID 47Google Scholar
  10. Li C-l, Liou Y-C, Yao Y: A damped algorithm for the split feasibility and fixed point problems. J. Inequal. Appl. 2013., 2013: Article ID 379Google Scholar
  11. Moudafi A: The split common fixed-point problem for demicontractive mappings. Inverse Probl. 2010., 26: Article ID 055007Google Scholar
  12. Ceng L-C, Petrusel A, Yao J-C: Strong convergence of modified implicit iterative algorithms with perturbed mappings for continuous pseudocontractive mappings. Appl. Math. Comput. 2009, 209: 162–176. 10.1016/j.amc.2008.10.062View ArticleMathSciNetGoogle Scholar
  13. Chen R, Song Y, Zhou H: Convergence theorems for implicit iteration process for a finite family of continuous pseudocontractive mappings. J. Math. Anal. Appl. 2006, 314: 701–709. 10.1016/j.jmaa.2005.04.018View ArticleMathSciNetGoogle Scholar
  14. Chidume CO, Souza GD: Convergence of a Halpern-type iteration algorithm for a class of pseudo-contractive mappings. Nonlinear Anal. 2008, 69: 2286–2292. 10.1016/j.na.2007.08.008View ArticleMathSciNetGoogle Scholar
  15. Morales CH, Jung JS: Convergence of paths for pseudocontractive mappings in Banach spaces. Proc. Am. Math. Soc. 2000, 128: 3411–3419. 10.1090/S0002-9939-00-05573-8View ArticleMathSciNetGoogle Scholar
  16. Yao Y, Liou Y-C, Chen R: Strong convergence of an iterative algorithm for pseudocontractive mappings in Banach spaces. Nonlinear Anal. 2007, 67: 3311–3317. 10.1016/j.na.2006.10.013View ArticleMathSciNetGoogle Scholar
  17. Zhou H: Strong convergence of an explicit iterative algorithm for continuous pseudo-contractions in Banach spaces. Nonlinear Anal. 2009, 70: 4039–4046. 10.1016/j.na.2008.08.012View ArticleMathSciNetGoogle Scholar
  18. Schu J: Approximating fixed points of Lipschitzian pseudocontractive mappings. Houst. J. Math. 1993, 19: 107–115.MathSciNetGoogle Scholar
  19. Udomene A: Path convergence, approximation of fixed points and variational solutions of Lipschitz pseudocontractions in Banach spaces. Nonlinear Anal. 2007, 67: 2403–2414. 10.1016/j.na.2006.09.001View ArticleMathSciNetGoogle Scholar
  20. Song Y: Strong convergence of viscosity approximation methods with strong pseudocontraction for Lipschitz pseudocontractive mappings. Positivity 2009, 13: 643–655. 10.1007/s11117-008-2246-3View ArticleMathSciNetGoogle Scholar
  21. Song Y: A note on the paper ‘A new iteration process for generalized Lipschitz pseudocontractive and generalized Lipschitz accretive mappings’. Nonlinear Anal. 2008, 68: 3047–3049. 10.1016/j.na.2007.02.044View ArticleMathSciNetGoogle Scholar
  22. Ishikawa S: Fixed point by a new iteration method. Proc. Am. Math. Soc. 1974, 4(1):147–150.View ArticleMathSciNetGoogle Scholar
  23. He Z: A new iterative scheme for equilibrium problems and fixed point problems of strict pseudo-contractive mappings and its application. Math. Commun. 2012, 17: 411–422.MathSciNetGoogle Scholar
  24. Yao Y, Postolache M, Liou Y-C: Coupling Ishikawa algorithms with hybrid techniques for pseudocontractive mappings. Fixed Point Theory Appl. 2013., 2013: Article ID 211Google Scholar
  25. Du, W-S, He, Z: Feasible iterative algorithms for split common solution problems. J. Nonlinear Convex Anal. (in press)Google Scholar
  26. Zhou H: Convergence theorems of fixed points for Lipschitz pseudo-contractions in Hilbert spaces. J. Math. Anal. Appl. 2008, 343: 546–556. 10.1016/j.jmaa.2008.01.045View ArticleMathSciNetGoogle Scholar
  27. Tang YC, Peng JG, Liu LW: Strong convergence theorem for pseudocontractive mappings in Hilbert spaces. Nonlinear Anal. 2011, 74: 380–385. 10.1016/j.na.2010.08.048View ArticleMathSciNetGoogle Scholar
  28. Chidume CE, Zegeye H: Approximate fixed point sequences and convergence theorems for Lipschitz pseudo-contractive maps. Proc. Am. Math. Soc. 2003, 132: 831–840.View ArticleMathSciNetGoogle Scholar
  29. Shahzad N, Zegeye H: Approximating a common point of fixed points of a pseudocontractive mapping and zeros of sum of monotone mappings. Fixed Point Theory Appl. 2014., 2014: Article ID 85Google Scholar

Copyright

© He and Du; licensee Springer. 2014

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.

Advertisement