Next Article in Journal
Geodesics and Translation Curves in Sol04
Previous Article in Journal
On a Conjecture of Cai–Zhang–Shen for Figurate Primes
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Approximate Solutions of a Fixed-Point Problem with an Algorithm Based on Unions of Nonexpansive Mappings

by
Alexander J. Zaslavski
Department of Mathematics, Technion–Israel Institute of Technology, Haifa 3200003, Israel
Mathematics 2023, 11(6), 1534; https://doi.org/10.3390/math11061534
Submission received: 14 February 2023 / Revised: 3 March 2023 / Accepted: 20 March 2023 / Published: 22 March 2023

Abstract

:
In this paper, we study a fixed-point problem with a set-valued mapping by using an algorithm based on unions of nonexpansive mappings. We show that an approximate solution is reached after a finite number of iterations in the presence of computational errors. This result is an extension of the results known in the literature.

1. Introduction

The study of fixed-point problems is an important topic in nonlinear analysis [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]. These problems have various applications in mathematical analysis, optimization theory, engineering, medicine, and the natural sciences [14,15,16,17,18,19,20]. In particular, in [21], a novel framework for the investigation of iterative algorithms was introduced. This framework was given in terms of a certain nonlinear set-valued map T defined on a space X. For every  x X T ( x )  is a finite union of values of single-valued paracontracting operators. Tam [21] established a convergence for this algorithm. Note that his result was a generalization of the result attained by Bauschke and Noll [22]. In our recent paper [23], we obtained an extension of a result of [21]. It should be mentioned that in [21], X is a finite-dimensional Euclidean space, while in [23] and in the present paper, X is an arbitrary metric space. The main result of [23] was obtained for inexact iterations of operators under the assumption that the common fixed-point problem has a solution. In the present paper, we prove an extension of this result in a case in which the common fixed-point problem has only an approximated solution.

2. Preliminaries

Assume that  ( X , ρ )  is a metric space endowed with a metric  ρ  and that  C X  is its nonempty closed set. For every  u X  and every  Δ ( 0 , ) , we set
B ( u , Δ ) = { v X : ρ ( u , v ) Δ } .
For every map  A : C C , we define
Fix ( A ) = { u C : A ( u ) = u } .
Assume that  T i : C C i = 1 , , m , where  m 1  is an integer,  0 < c ¯ 1 , and that for every  j { 1 , , m } , every  u Fix ( T j ) , and every  v C ,
ρ ( u , v ) 2 ρ ( u , T j ( v ) ) 2 c ¯ ρ ( v , T j ( v ) ) 2 .
It should be mentioned that inequality (1) is true for many nonlinear operators [14,15].
Assume that
ϕ : X 2 { 1 , , m } \ { } .
We set
T ( u ) = { T j ( u ) : j ϕ ( u ) } .
for each  u C  and
F ( T ) = { u C : u T ( u ) } .
In this paper, we study the fixed-point problem
Find x X such   that x T ( x ) .
This problem was introduced and studied in [21]. It should be mentioned that in [21], X was a finite-dimensional Euclidean space, and the mappings  T i i = 1 , , m  were paracontracting. Tam [21] considered a sequence of iterations  { x k } k = 0 X  satisfying  x k + 1 T ( x k )  for every integer  k 0  and established its convergence under the assumption that the mappings  T i i = 1 , , m  had a common fixed point. In [21], this convergent result was applied to sparsity-constrained minimization. Note that the result in [21] was a generalization of the result attained by Bauschke and Noll [22]. In our recent paper [23], we considered mappings acting on a general metric space and obtained two extensions of the result from [21]. In the first result, we studied exact iterations of the set-valued mapping, while in the second one, we dealt with its inexact iterations while taking computational errors into account. More precisely, in [23], for a given computational error  δ > 0 , we considered a sequence  { x k } k = 0 X  satisfying  B ( x k + 1 , δ ) T ( x k )  for every integer  k 0  and analyzed its behavior. This result was also obtained under the assumption that the mappings  T i i = 1 , , m  had a common fixed point. In the present paper, we generalize this result. Instead of assuming the existence of a common fixed point, we suppose that there exists an approximate common fixed point z such that
B ( z , γ ) Fix ( T i ) , i = 1 , , m ,
where  γ  is a given small positive constant. In other words, a small neighborhood of z contains a fixed point of every mapping.
We fix
θ C .
For any  u R 1 , we set
u = max { j : j as   an   integer   and j u } .
We prove the following theorem in the presence of computational errors. This theorem shows that after a certain number of iterations, we obtain an approximate solution to our fixed-point problem. The number of iterations depends on the computational error.
Theorem 1. 
Let  M > 0 ϵ ( 0 , 1 ] ,
γ ( 0 , ( 18 ) 1 ( 4 M + 4 ) 1 ϵ 2 c ¯ ) ,
z B ( θ , M )
satisfy
B ( z , γ ) Fix ( T i ) , i = 1 , , m ,
Q = 8 ϵ 2 M 2 c ¯ 1 + 1 ,
and  δ ( 0 , γ ) .  Assume that  { x k } k = 0 C ,
ρ ( θ , x 0 ) M
and that
B ( x k + 1 , δ ) T ( x k ) , k = 0 , 1 , .
Then, there is a nonnegative integer  p < Q  for which
B ( x p , ϵ ) T ( x p ) .
In the theorem above, we assume the existence of a point z that satisfies (7), which means that z is an approximate fixed point for all of the mappings  T i i = 1 , , m . This result has a prototype in [23], which was obtained under the assumption that z is a common fixed point for all  T k k = 1 , , m .

3. Proof of Theorem 1

Proof. 
Assume that for every nonnegative integer  k < Q , relation (11) is not true. Then, for every nonnegative integer  k < Q ,
B ( x k , ϵ ) T ( x k ) = .
We set
M 0 = 2 M + 1 .
According to (7), for every  k { 1 , , m } , there is
z k Fix ( T k )
such that
ρ ( z , z k ) γ .
According to (6) and (9),
ρ ( x 0 , z ) 2 M .
Let  i [ 0 , Q 1 ]  be an integer. According to (10), there is
x ^ i + 1 T ( x i )
for which
ρ ( x i + 1 , x ^ i + 1 ) δ .
Equations (3) and (17) imply that there is an integer  j [ 1 , m ]  for which
x ^ i + 1 = T j ( x i ) .
It follows from (1) and (19) that
ρ ( z j , x i ) 2 ρ ( z j , x ^ i + 1 ) 2 + c ¯ ρ ( x i , x ^ i + 1 ) 2 .
According to (12) and (19),
ρ ( x i , x ^ i + 1 ) > ϵ .
In view of (20) and (21),
ρ ( z j , x i ) 2 ρ ( z j , x ^ i + 1 ) 2 + c ¯ ϵ 2 .
Assume that
ρ ( z , x i ) M 0 .
(In view of (13) and (16), Equation (23) holds for  i = 0 ). Equations (15) and (23) imply that
ρ ( z j , x i ) ρ ( z j , z ) + ρ ( z , x i ) M 0 + γ .
It follows from (5), (13), (22), and (24) that
ρ ( z j , x ^ i + 1 ) 2 ρ ( z j , x i ) 2 ϵ 2 c ¯ ( M 0 + γ ) 2 ϵ 2 c ¯
= M 0 2 + γ ( γ + 2 M 0 ) ϵ 2 c ¯ M 0 2 + γ ( 1 + 2 M 0 ) ϵ 2 c ¯
M 0 2 7 γ ( 1 + 2 M 0 ) ( M 0 2 γ ) 2
and
ρ ( z j , x ^ i + 1 ) M 0 2 γ .
According to (15) and (25),
ρ ( z , x i + 1 ) ρ ( z , z j ) + ρ ( z j , x ^ t + 1 ) + ρ ( x ^ i + 1 , x i + 1 )
M 0 2 γ + γ + γ M 0
and
ρ ( z , x i + 1 ) M 0 .
According to (22),
ρ ( z j , x ^ i + 1 ) 2 ρ ( z j , x i ) 2 ϵ 2 c ¯ .
Equations (15) and (23) imply that
| ρ ( x i , z ) 2 ρ ( x i , z j ) 2 |
( ρ ( x i , z ) + ρ ( x i , z j ) ) | ρ ( x i , z ) ρ ( x i , z j ) |
( ρ ( x i , z ) + ρ ( x i , z ) + γ ) ρ ( z j , z ) γ ( 2 M 0 + 1 ) .
It follows from (15), (18), and (26) that
| ρ ( x i + 1 , z ) 2 ρ ( x ^ i + 1 , z j ) 2 |
( ρ ( x i + 1 , z ) + ρ ( x ^ i + 1 , z j ) ) | ρ ( x i + 1 , z ) ρ ( x ^ i + 1 , z j ) |
( 2 M 0 + γ + δ ) ( ρ ( z j , z ) + ρ ( x ^ i + 1 , x i + 1 ) ) ( 2 M 0 + 2 ) ( γ + δ ) .
By (5), (13), (22), and (29),
ρ ( x i + 1 , z ) 2 ρ ( z j , x ^ i + 1 ) 2 + 2 γ ( 2 M 0 + 2 )
ρ ( z j , x i ) 2 c ¯ ϵ 2 + 2 γ ( 2 M 0 + 2 )
ρ ( x i , z ) 2 ϵ 2 c ¯ + γ ( 2 M 0 + 1 ) + 2 γ ( 2 M 0 + 2 )
ρ ( x i , z ) 2 ϵ 2 c ¯ + 3 γ ( 2 M 0 + 1 )
ρ ( x i , z ) 2 ϵ 2 c ¯ / 2 .
Thus, we have shown by induction that (23) and (30) hold for  i = 0 , , Q 1 . By (16) and (30),
4 M 2 ρ ( z , x 0 ) ) 2
ρ ( z , x 0 ) 2 ρ ( z , x Q ) 2
= i = 0 Q 1 ( ρ ( z , x i ) 2 ρ ( z , x i + 1 ) 2 ) Q c ¯ ϵ 2 / 2 ,
and
Q 8 M 2 c ¯ 1 ϵ 2 .
This contradicts (8). The contradiction that we have reached proves Theorem 1. □

4. Extensions

We use the notation and definitions introduced in Section 2.
Lemma 1. 
Assume that  M 0 > 0 ,
z B ( θ , M 0 ) ,
B ( z , 1 ) Fix ( T i ) , i = 1 , , m ,
x 0 B ( θ , M 0 ) ,
x 1 C , and
B ( x 1 , 1 ) T ( x 0 ) .
Then,
ρ ( x 1 , θ ) 3 M 0 + 3 .
Proof. 
According to (3), there is an integer  j [ 1 , m ]  for which
ρ ( x 1 , T j ( x 0 ) ) 1 .
According to (32), there is
z j Fix ( T j )
for which
ρ ( z , z j ) 1 .
Equations (1), (31), (33), and (35)–(37) imply that
ρ ( x 1 , θ ) ρ ( θ , T j ( x 0 ) ) + ρ ( T j ( x 0 ) , x 1 )
1 + ρ ( θ , z ) + ρ ( z , z j ) + ρ ( z j , T j ( x 0 ) )
1 + M 0 + 1 + ρ ( z j , x 0 )
2 + M 0 + ρ ( θ , x 0 ) + ρ ( θ , z ) + ρ ( z , z j )
3 + 3 M 0 .
Lemma 1 is proved. □
Theorem 2. 
Let  M > 0 ϵ ( 0 , 1 ] ,
γ ( 0 , ( 18 ) 1 ( 12 M + 12 ) 1 ϵ 2 c ¯ ) ,
z B ( θ , M )
satisfy
B ( z , γ ) Fix ( T i ) , i = 1 , , m ,
Q = 8 ϵ 2 ( 3 M + 3 ) 2 c ¯ 1 + 1 ,
and  δ ( 0 , γ ) .
Assume that  { x k } k = 0 C ,
ρ ( θ , x 0 ) M ,
and that
B ( x k + 1 , δ ) T ( x k ) , k = 0 , 1 , .
Then, there is  j { 1 , , Q }  for which
B ( x j , ϵ ) T ( x j ) .
Proof. 
Lemma 1 implies that
ρ ( x 1 , θ ) 3 M + 3 .
The application of Theorem 1 to the sequence  { x i + 1 } i = 0  implies our result. □
Theorem 3. 
Let  M > 0 ϵ ( 0 , 1 ] ,
{ ξ C : B ( ξ , ϵ ) T ( ξ ) } B ( θ , M ) ,
γ ( 0 , ( 18 ) 1 ( 12 M + 12 ) 1 ϵ 2 c ¯ ) ,
z B ( θ , M )
satisfy
B ( z , γ ) Fix ( T i ) , i = 1 , , m ,
Q = 8 ϵ 2 ( 3 M + 3 ) 2 c ¯ 1 + 1 ,
and  δ ( 0 , γ ) .
Assume that  { x k } k = 0 C ,
ρ ( θ , x 0 ) M ,
and that
B ( x k + 1 , δ ) T ( x k ) , k = 0 , 1 , .
Then, there exists a strictly increasing sequence of natural numbers  { q j } j = 1  such that
1 q 0 Q ,
and for each integer  j 0 ,
q j + 1 q j Q
B ( x q j , ϵ ) T ( x q j ) .
Proof. 
Theorem 2 implies that there exists  q 0 { 1 , , Q }  for which
B ( x q 0 , ϵ ) T ( x q 0 ) .
Assume that  p { 0 , 1 , } q j j = 0 , , p  are natural numbers such that for any integer j satisfying  0 j < p , (40) holds, and assume that (41) is true for all  j = 0 , , p . We set
y i = x i + q p , i = 0 , 1 , .
According to (38) and (41),
ρ ( θ , y 0 ) M .
Clearly, all of the assumptions of Theorem 2 hold with  x i = y i i = 0 , 1 , , and Theorem 2 implies that there is  j { 1 , , Q }  for which
B ( y j , ϵ ) T ( y j ) .
We set
q p + 1 = q p + j .
Clearly,
B ( x q p + 1 , ϵ ) T ( x q p + 1 ) .
Thus, by induction, we have constructed the sequence of natural numbers  { q j } j = 1  and proved Theorem 3. □

Funding

This research received no external funding.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Bejenaru, A.; Postolache, M. An unifying approach for some nonexpansiveness conditions on modular vector spaces. Nonlinear Anal. Model. Control 2020, 25, 827–845. [Google Scholar] [CrossRef]
  2. Goebel, K.; Kirk, W.A. Topics in Metric Fixed Point Theory; Cambridge University Press: Cambridge, UK, 1990. [Google Scholar]
  3. Goebel, K.; Reich, S. Uniform Convexity, Hyperbolic Geometry, and Nonexpansive Mappings; Marcel Dekker: New York, NY, USA; Basel, Switzerland, 1984. [Google Scholar]
  4. Kassab, W.; Turcanu, T. Numerical reckoning fixed points of (E)-type mappings in modular vector spaces. Mathematics 2019, 7, 390. [Google Scholar] [CrossRef] [Green Version]
  5. Khamsi, M.A.; Kozlowski, W.M. Fixed Point Theory in Modular Function Spaces; Birkhäser/Springer: Cham, Switzerland, 2015. [Google Scholar]
  6. Khamsi, M.A.; Kozlowski, W.M.; Reich, S. Fixed Point Theory in Modular Function Spaces. Nonlinear Anal. 1990, 14, 935–953. [Google Scholar] [CrossRef] [Green Version]
  7. Kirk, W.A. Contraction mappings and extensions. In Handbook of Metric Fixed Point Theory; Kluwer: Dordrecht, The Netherlands, 2001; pp. 1–34. [Google Scholar]
  8. Kozlowski, W.M. An introduction to fixed point theory in modular function spaces. In Topics in Fixed Point Theory; Springer: Cham, Switzerland, 2014; pp. 15–222. [Google Scholar]
  9. Kubota, R.; Takahashi, W.; Takeuchi, Y. Extensions of Browder’s demiclosedness principle and Reich’s lemma and their applications. Pure Appl. Func. Anal. 2016, 1, 63–84. [Google Scholar]
  10. Okeke, G.A.; Abbas, M.; de la Sen, M. Approximation of the fixed point of multivalued quasi-nonexpansive mappings via a faster iterative process with applications. Discret. Dyn. Nat. Soc. 2020, 2020, 8634050. [Google Scholar] [CrossRef]
  11. Okeke, G.A.; Ugwuogor, C.I. Iterative construction of the fixed point of Suzukis generalized nonexpansive mappings in Banach spaces. Fixed Point Theory 2022, 23, 633–652. [Google Scholar]
  12. Rakotch, E. A note on contractive mappings. Proc. Am. Math. Soc. 1962, 13, 459–465. [Google Scholar] [CrossRef]
  13. Reich, S.; Zaslavski, A.J. Genericity in nonlinear analysis. In Developments in Mathematics; Springer: New York, NY, USA, 2014; Volume 34. [Google Scholar]
  14. Zaslavski, A.J. Approximate solutions of common fixed point problems. In Springer Optimization and Its Applications; Springer: Cham, Switzerland, 2016. [Google Scholar]
  15. Zaslavski, A.J. Algorithms for solving common fixed point problems. In Springer Optimization and Its Applications; Springer: Cham, Switzerland, 2018. [Google Scholar]
  16. Censor, Y.; Zaknoon, M. Algorithms and convergence results of projection methods for inconsistent feasibility problems: A review. Pure Appl. Func. Anal. 2018, 3, 565–586. [Google Scholar]
  17. Gibali, A. A new split inverse problem and an application to least intensity feasible solutions. Pure Appl. Funct. Anal. 2017, 2, 243–258. [Google Scholar]
  18. Gibali, A.; Reich, S.; Zalas, R. Outer approximation methods for solving variational inequalities in Hilbert space. Optimization 2017, 66, 417–437. [Google Scholar] [CrossRef] [Green Version]
  19. Takahashi, W. The split common fixed point problem and the shrinking projection method for new nonlinear mappings in two Banach spaces. Pure Appl. Funct. Anal. 2017, 2, 685–699. [Google Scholar]
  20. Takahashi, W. A general iterative method for split common fixed point problems in Hilbert spaces and applications. Pure Appl. Funct. Anal. 2018, 3, 349–369. [Google Scholar]
  21. Tam, M.K. Algorithms based on unions of nonexpansive maps. Optim. Lett. 2018, 12, 1019–1027. [Google Scholar] [CrossRef] [Green Version]
  22. Bauschke, H.H.; Noll, D. On the local convergence of the Douglas-Rachford algorithm. Arch. Math. 2014, 102, 589–600. [Google Scholar] [CrossRef] [Green Version]
  23. Zaslavski, A.J. An algorithm based on unions of nonexpansive mappings in metric spaces. Symmetry 2022, 14, 1852. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zaslavski, A.J. Approximate Solutions of a Fixed-Point Problem with an Algorithm Based on Unions of Nonexpansive Mappings. Mathematics 2023, 11, 1534. https://doi.org/10.3390/math11061534

AMA Style

Zaslavski AJ. Approximate Solutions of a Fixed-Point Problem with an Algorithm Based on Unions of Nonexpansive Mappings. Mathematics. 2023; 11(6):1534. https://doi.org/10.3390/math11061534

Chicago/Turabian Style

Zaslavski, Alexander J. 2023. "Approximate Solutions of a Fixed-Point Problem with an Algorithm Based on Unions of Nonexpansive Mappings" Mathematics 11, no. 6: 1534. https://doi.org/10.3390/math11061534

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop