Share This Article:

Results on Generalized Quasi Contraction Random Operators

DOI: 10.4236/oalib.1103539    600 Downloads   877 Views   Citations

ABSTRACT

In this paper, we prove the existence of common random fixed point for two random operators under general quasi contraction condition in a complete p-normed space X (with whose dual separates the point of X). Also, the well-posedness problem of random fixed points is studied. Our results essentially cover special cases.

1. Introduction

Let X be a linear space and p be a real valued function on X with 0 < p 1 .

The ordered pair ( X , p ) is called a p-normed space [1] if for all x , y in X and scalars λ :

1) x p 0 and x p = 0 iff x = 0

2) λ x p = | λ | p x p

3) x + y p x p + y p

for more details about p-normed spaces, see [2] or [3] . Throughout this article, X will be complete p-normed space whose dual separates the points of it, A X be a separable closed, ( Ω , Σ ) be the measurable space with Σ a sigma algebra of subsets of Ω.

Definition (1.1): [4]

A mapping F : Ω X is called measurable if, for open subset B of,

F 1 ( B ) = { γ Ω : F ( γ ) B } Σ .

Definition (1.2): [4]

A mapping h : Ω × X X is called a random operator if for any x X , h ( . , x ) is measurable.

Definition (1.3): [5]

A measurable mapping λ : Ω A is called random fixed point of a random operator h : Ω × X X if for every γ Ω , λ ( γ ) = h ( γ , λ ( γ ) ) .

Definition (1.4): [6]

A measurable mapping λ : Ω A is called common random fixed point of a random operator h : Ω × A X and G : Ω × A A if for all γ Ω

λ ( γ ) = h ( γ , λ ( γ ) ) = G ( γ , λ ( γ ) ) .

Definition (1.5): [7]

A random operator h : Ω × A X is called continuous (weakly continuous) if for each γ Ω , h ( γ , . ) is continuous (weakly continuous).

The stochastic generalization of fixed point theory is random fixed point theory. Many researchers are interesting in this subject and it’s applications in best approximations, integral equations and differential equations such as [8] - [14] .

Saluj [15] establish some common random fixed point theorems under contractive type condition in the framework of cone random metric spaces. Rashwan and Albaqeri [16] obtained common random fixed point theorems for six weakly compatible random operators defined on a nonempty closed subset of a separable Hilbert space. In 2013, Arunchaiand Plubtieng [17] proved some random fixed point theorem for some of weakly-strongly continuous random operators and nonexpansive random operators in Banach spaces. Singh, Rathore, Dubey and Singh [18] obtain a common random fixed point theorem for four continuous random operators in separable Hilbert spaces. Vishwakarme and Chauhan [19] proved common random fixed point theorems for weakly compatible random operators in symmetric spaces. Khanday, Jain and Badshah [20] proved the existence of common random fixed point theorems of two random multivalued generalized contractions by using functional expressions. Chanhan [21] obtained common random fixed point theorems for four continuous random operators satisfying certain contractive conditions in separable Hilbert spaces. In 2017 Abed, Ajeel, and Alsaidy [22] proved the existence of common random fixed point for two continuous random operators under quasi contraction condition in a complete p-normed space X. Also, the random coincidence point results are proved in [23] for -weakly contraction condition under two pairs of random operators.

Now, we define a new type of random operators

Definition (1.6):

Let A be a nonempty subset of a p-normed space, let ( Ω , Σ ) be a measurable space and let h , G : Ω × A A be tow random operators. The random operator h is called

1. Generalized quasi contraction (gqc) random operator if for each γ Ω , the mapping h ( γ , . ) : A A satisfies the following condition

h ( γ , x ) h ( γ , y ) p k max { x y p , x h ( γ , x ) p , y h ( γ , y ) p , x h ( γ , y ) p , y h ( γ , x ) p , h 2 ( γ , x ) x p , h 2 ( γ , x ) h ( γ , x ) p , h 2 ( γ , x ) y p , h 2 ( γ , x ) h ( γ , y ) p } (1.1)

For all x , y A and 0 k < 1 / 2 .

2. G-generalized quasi contraction (G-gqc) random operator if for each γ Ω , the mappings h ( γ , . ) , G ( γ , . ) : A A satisfies the following condition

h ( γ , x ) G ( γ , y ) p k max { x y p , x h ( γ , x ) p , y G ( γ , y ) p , x G ( γ , y ) p , y h ( γ , x ) p , h 2 ( γ , x ) x p , h 2 ( γ , x ) h ( γ , x ) p , h 2 ( γ , x ) y p , h 2 ( γ , x ) G ( γ , y ) p } (1.2)

For all x , y A and 0 k < 1 / 2 .

2. Common Random Fixed Point Theorems

We begin with the following result

Theorem (2.1):

Let A X , G : Ω × A A be a continuous random operator and h : Ω × A A be a nonexpansive random operator. If A be a separable closed subset of a complete p-Normed space X and h be G-gqc random operator, then h and G have a unique common random fixed point.

Proof:

Let λ ° : Ω A be arbitrary measurable mapping. We construct a sequence of measurable mappings λ n on Ω to A as follows

Let λ 1 , λ 2 : Ω A be tow measurable mappings such that

h ( γ , λ ° ( γ ) ) = λ 1 ( γ ) and G ( γ , λ 1 ( γ ) ) = λ 2 ( γ )

By induction, we construct sequence of measurable mappings λ n : Ω A such that

h ( γ , λ 2 n 1 ( γ ) ) = λ 2 n ( γ ) and G ( γ , λ 2 n ( γ ) ) = λ 2 n + 1 ( γ ) (2.1)

From (2.1) and (1.2), we have

λ 2 n ( γ ) λ 2 n + 1 ( γ ) p = h ( γ , λ 2 n 1 ( γ ) ) G ( γ , λ 2 n ( γ ) ) p k max { λ 2 n 1 ( γ ) λ 2 n ( γ ) p , λ 2 n 1 ( γ ) h ( γ , λ 2 n 1 ( γ ) ) p , λ 2 n ( γ ) G ( γ , λ 2 n ( γ ) ) p , λ 2 n 1 ( γ ) G ( γ , λ 2 n ( γ ) ) p , λ 2 n ( γ ) h ( γ , λ 2 n 1 ( γ ) ) p , h 2 ( γ , λ 2 n 1 ( γ ) ) λ 2 n 1 ( γ ) p , h 2 ( γ , λ 2 n 1 ( γ ) ) h ( γ , λ 2 n 1 ( γ ) ) p , h 2 ( γ , λ 2 n 1 ( γ ) ) λ 2 n ( γ ) p , h 2 ( γ , λ 2 n 1 ( γ ) ) G ( γ , λ 2 n ( γ ) ) p } = k max { λ 2 n 1 ( γ ) λ 2 n ( γ ) p , λ 2 n 1 ( γ ) λ 2 n ( γ ) p , λ 2 n ( γ ) λ 2 n + 1 ( γ ) p , λ 2 n 1 ( γ ) λ 2 n + 1 ( γ ) p , λ 2 n ( γ ) λ 2 n ( γ ) p , λ 2 n + 1 ( γ ) λ 2 n 1 ( γ ) p , λ 2 n + 1 ( γ ) λ 2 n ( γ ) p , λ 2 n + 1 ( γ ) λ 2 n ( γ ) p , λ 2 n + 1 ( γ ) λ 2 n + 1 ( γ ) p } = k max { λ 2 n 1 ( γ ) λ 2 n ( γ ) p , λ 2 n ( γ ) λ 2 n + 1 ( γ ) p , λ 2 n 1 ( γ ) λ 2 n + 1 ( γ ) p }

using triangle inequality, we get

λ 2 n ( γ ) λ 2 n + 1 ( γ ) p k max { λ 2 n 1 ( γ ) λ 2 n ( γ ) p , λ 2 n ( γ ) λ 2 n + 1 ( γ ) p , λ 2 n 1 ( γ ) λ 2 n ( γ ) p + λ 2 n ( γ ) λ 2 n + 1 ( γ ) p } = k [ λ 2 n 1 ( γ ) λ 2 n ( γ ) p + λ 2 n ( γ ) λ 2 n + 1 ( γ ) p ]

hence,

λ 2 n ( γ ) λ 2 n + 1 ( γ ) p λ λ 2 n 1 ( γ ) λ 2 n ( γ ) p

where λ = ( k / 1 k ) < 1 .

By similar way, we have

λ 2 n 1 ( γ ) λ 2 n ( γ ) p λ λ 2 n 2 ( γ ) λ 2 n 1 ( γ ) p

therefore,

λ 2 n ( γ ) λ 2 n + 1 ( γ ) p λ λ 2 n 1 ( γ ) λ 2 n ( γ ) p λ 2 λ 2 n 2 ( γ ) λ 2 n 1 ( γ ) p λ 2 n ( γ ) λ 2 n + 1 ( γ ) p λ 2 n λ 0 ( γ ) λ 1 ( γ ) p

To prove λ n is Cauchy sequence, for n , m , n > m

λ n ( γ ) λ m ( γ ) p λ n ( γ ) λ n 1 ( γ ) p + λ n 1 ( γ ) λ n 2 ( γ ) p + + λ m + 1 ( γ ) λ m ( γ ) p ( λ n 1 + λ n 2 + + λ m ) λ 0 ( γ ) λ 1 ( γ ) p ( λ m / 1 λ ) λ 0 ( γ ) λ 1 ( γ ) p

Let ϵ > 0 be given, choose a natural number K large enough such that λ m λ 1 ( γ ) λ 0 ( γ ) p < ϵ for every m K .

Hence λ n ( γ ) λ m ( γ ) p < ϵ for every n > m K .

So, { λ n ( γ ) } is a Cauchy sequence in, and completeness of X implise that there exists λ ( γ ) X such that λ n ( γ ) λ ( γ ) as n .

To show that λ is a common random fixed point of h and G, coinsider the following by using triangle inequality, (2.1) and (1.2)

λ ( γ ) h ( γ , λ ( γ ) ) p λ ( γ ) λ 2 n + 2 ( γ ) p + λ 2 n + 2 ( γ ) h ( γ , λ ( γ ) ) p = λ ( γ ) λ 2 n + 2 ( γ ) p + h ( γ , λ ( γ ) ) G ( γ , λ 2 n + 1 ( γ ) ) p λ ( γ ) λ 2 n + 2 ( γ ) p + k max { λ ( γ ) λ 2 n + 1 ( γ ) p , λ ( γ ) h ( γ , λ ( γ ) ) p , λ 2 n + 1 ( γ ) G ( γ , λ 2 n + 1 ( γ ) ) p , λ ( γ ) G ( γ , λ 2 n + 1 ( γ ) ) p , λ 2 n + 1 ( γ ) h ( γ , λ ( γ ) ) p , h 2 ( γ , λ ( γ ) ) λ ( γ ) p , h 2 ( γ , λ ( γ ) ) h ( γ , λ ( γ ) ) p , h 2 ( γ , λ ( γ ) ) λ 2 n + 1 ( γ ) p , h 2 ( γ , λ ( γ ) ) G ( γ , λ 2 n + 1 ( γ ) ) p } = λ ( γ ) λ 2 n + 2 ( γ ) p + k max { λ ( γ ) λ 2 n + 1 ( γ ) p , λ ( γ ) h ( γ , λ ( γ ) ) p , λ 2 n + 1 ( γ ) λ 2 n + 2 ( γ ) p , λ ( γ ) λ 2 n + 2 ( γ ) p , λ 2 n + 1 ( γ ) h ( γ , λ ( γ ) ) p , h 2 ( γ , λ ( γ ) ) λ ( γ ) p , h 2 ( γ , λ ( γ ) ) h ( γ , λ ( γ ) ) p , h 2 ( γ , λ ( γ ) ) λ 2 n + 1 ( γ ) p , h 2 ( γ , λ ( γ ) ) λ 2 n + 2 ( γ ) p }

taking the limit as n in the above inequality, getting that

λ ( γ ) h ( γ , λ ( γ ) ) p k max { λ ( γ ) h ( γ , λ ( γ ) ) p , h 2 ( γ , λ ( γ ) ) λ ( γ ) p , h 2 ( γ , λ ( γ ) ) h ( γ , λ ( γ ) ) p }

By using triangle inequality and non-expansive of h, we have

λ ( γ ) h ( γ , λ ( γ ) ) p k max { λ ( γ ) h ( γ , λ ( γ ) ) p , h 2 ( γ , λ ( γ ) ) h ( γ , λ ( γ ) ) p + h ( γ , λ ( γ ) ) λ ( γ ) p , h 2 ( γ , λ ( γ ) ) h ( γ , λ ( γ ) ) p } k max { λ ( γ ) h ( γ , λ ( γ ) ) p , h ( γ , λ ( γ ) ) λ ( γ ) p + h ( γ , λ ( γ ) ) λ ( γ ) p , h ( γ , λ ( γ ) ) λ ( γ ) p } 2 k h ( γ , λ ( γ ) ) λ ( γ ) p

this implies that

( 1 k ) λ ( γ ) h ( γ , λ ( γ ) ) p 0 (2.2)

since 0 k < 1 / 2 , (2.1.4) must be true only λ ( γ ) h ( γ , λ ( γ ) ) p = 0 , thus

λ ( γ ) = h ( γ , λ ( γ ) ) (2.3)

Similarly, we can show that

λ ( γ ) = G ( γ , λ ( γ ) ) (2.4)

hence λ : Ω A is a common random fixed point of h and G.

For uniqueness, let α ( γ ) be another common random fixed point of S and T, that is for all γ Ω , α ( γ ) = h ( γ , α ( γ ) ) = G ( γ , α ( γ ) ) .

Then for all γ Ω , we have

λ ( γ ) α ( γ ) p = h ( γ , λ ( γ ) ) G ( γ , α ( γ ) ) p

From (1.2), (2.3) and (3.2), we have

λ ( γ ) α ( γ ) p k max { λ ( γ ) α ( γ ) p , λ ( γ ) h ( γ , λ ( γ ) ) p , α ( γ ) T ( γ , α ( γ ) ) p , λ ( γ ) G ( γ , α ( γ ) ) p p , α ( γ ) h ( γ , λ ( γ ) ) p , h 2 ( γ , λ ( γ ) ) λ ( γ ) p , h 2 ( γ , λ ( γ ) ) λ ( γ ) p , h 2 ( γ , λ ( γ ) ) α ( γ ) p , h 2 ( γ , λ ( γ ) ) G ( γ , α ( γ ) ) p } = k max { λ ( γ ) α ( γ ) p , 0 } = k λ ( γ ) α ( γ ) p < λ ( γ ) α ( γ ) p

Which is contraction. Hence λ : Ω A is a unique common random fixed point of h and. ∎

Corollary (2.2):

If A and h as in theorem (2.1) and for each γ Ω , h ( γ , . ) : A A is (gqc):

h ( γ , x ) h ( γ , y ) p k max { x y p , x h ( γ , x ) p , y h ( γ , y ) p , x h ( γ , y ) p , y h ( γ , x ) p , h 2 ( γ , x ) x p , h 2 ( γ , x ) h ( γ , x ) p , h 2 ( γ , x ) y p , h 2 ( γ , x ) h ( γ , y ) p }

Then there is a random fixed point of h.

Corollary (2.3):

If A, h, G as in theorem (2.1) and for each γ Ω , h ( γ , . ) , G ( γ , . ) : A A satisfies one of the following conditions:

1. h ( γ , x ) G ( γ , y ) p k max { x y p , x h ( γ , x ) p , y G ( γ , y ) p , x T ( γ , y ) p , y h ( γ , x ) p } .

2. h ( γ , x ) G ( γ , y ) p k max { x y p , x h ( γ , x ) p , y G ( γ , y ) p } .

3. h ( γ , x ) G ( γ , y ) p k max { x h ( γ , x ) p , y G ( γ , y ) p } .

4. h ( γ , x ) G ( γ , y ) p max { x y p , x h ( γ , x ) p , y G ( γ , y ) p , 1 / 2 [ x G ( γ , y ) p + y h ( γ , x ) p ] } .

5. h ( γ , x ) G ( γ , y ) p k max { x y p , 1 / 2 [ x h ( γ , x ) p + y G ( γ , y ) p ] , 1 / 2 [ x G ( γ , y ) p + y h ( γ , x ) p ] } .

For all x , y X ; 0 < k < 1 / 2 . Then h and G have a unique common random fixed point.

3. Well-Posed Problem

Definition (3.1):

Let ( X , p ) be a p-normed space and T : Ω × X X a random mapping. the random fixed point problem of T is said to be well-posed if:

i. T has a unique random fixed point λ : Ω X ;

ii. for any sequence { λ n ( γ ) } of measurable mappings in X such that lim n T ( γ , λ n ( γ ) ) λ n ( γ ) p = 0 , we have lim n λ n ( γ ) λ ( γ ) p = 0 .

Definition (3.2):

Let ( X , p ) be a p-normed space and let T be a set of a random operators in X. The random fixed point of T is said to be well-posed if:

i. T has a unique random fixed point λ : Ω X ;

ii. for any sequence { λ n ( γ ) } of measurable mappings in X such that lim n T ( γ , λ n ( γ ) ) λ n ( γ ) p = 0 , T T we have lim n λ n ( γ ) λ ( γ ) p = 0 .

Theorem (3.3):

If A, h, G as in theorem (2.1) and for each γ Ω , h ( γ , . ) , G ( γ , . ) : A A satisfies (1.2), then the common random fixed point for the set of random operators { h , G } is well-posed.

Proof:

By theorem (2.1), the random operators h and G have a unique common random fixed point λ : Ω A . Let { λ n ( γ ) } be a sequence of measurable mappings in A such that

lim n h ( γ , λ n ( γ ) ) λ n ( γ ) p = lim n G ( γ , λ n ( γ ) ) λ n ( γ ) p = 0

By the triangle inequality, (1.2), (2.3) and (2.4), we have

λ ( γ ) λ n ( γ ) p h ( γ , λ ( γ ) ) G ( γ , λ n ( γ ) ) p + G ( γ , λ n ( γ ) ) λ n ( γ ) p h max { λ ( γ ) λ n ( γ ) p , λ n ( γ ) G ( γ , λ n ( γ ) ) p , λ ( γ ) G ( γ , λ n ( γ ) ) p λ n ( γ ) h ( γ , λ ( γ ) ) p , h 2 ( γ , λ ( γ ) ) λ n ( γ ) p , h 2 ( γ , λ ( γ ) ) G ( γ , λ n ( γ ) ) p } + G ( γ , λ n ( γ ) ) λ n ( γ ) p h [ λ ( γ ) G ( γ , λ n ( γ ) ) p + G ( γ , λ n ( γ ) ) λ n ( γ ) p ] + G ( γ , λ n ( γ ) ) λ n ( γ ) p = h λ ( γ ) G ( γ , λ n ( γ ) ) p + ( 1 + h ) θ p

By the triangle inequality, we get

λ ( γ ) λ n ( γ ) p h [ λ ( γ ) λ n ( γ ) p + λ n ( γ ) G ( γ , λ n ( γ ) ) p ] + ( 1 + h ) G ( γ , λ n ( γ ) ) λ n ( γ ) p = h λ ( γ ) λ n ( γ ) p + ( 1 + 2 h ) G ( γ , λ n ( γ ) ) λ n ( γ ) p

( 1 h ) λ ( γ ) λ n ( γ ) p ( 1 + 2 h ) G ( γ , λ n ( γ ) ) λ n ( γ ) p

thus we have, lim n λ ( γ ) λ n ( γ ) p = 0 , it follows that the common random fixed point for the set of random operators { h , G } is well-posed. ∎

Conflicts of Interest

The authors declare no conflicts of interest.

Cite this paper

Abed, S. , Alsaidy, S. and Ajeel, Y. (2017) Results on Generalized Quasi Contraction Random Operators. Open Access Library Journal, 4, 1-9. doi: 10.4236/oalib.1103539.

References

[1] Nashine, H.K. (2006) Best Approximation for Nonconvex Set in q-Normed Space. Archivum Mathematicum (Brno), 42, 51-58.
[2] Bano, A., Khan, A.R. and Latif, A. (2003) Coincidence Points and Best Approximations in p-Normed Spaces. Radovi Matematicki, 12, 27-36.
[3] Lahiri, B.K. and Das, P. (2005) Well-Posednes and Porosity of Certain Classes of Operators. Demonstratio Mathematica, 38, 170-176.
[4] Jhadd, P.K. and Salua, A.S. (2014) On Random Coincidence and Fixed Points for a Pair of Multi-Valued and Single-Valued Mappings. International Journal of Analysis and Applications, 4, 26-35.
[5] Shahzad, N. and Latif, A. (2000) Note a Random Coincidence Point Theorem. Journal of Mathematical Analysis and Applications, 245, 633-638.
https://doi.org/10.1006/jmaa.2000.6772
[6] Beg, I., Jahangir, A. and Azam, A. (2009) Random Coincidence and Fixed Points for Weakly Compatible Mappings in Convex Metric Spaces. Asian-European Journal of Mathematics, 2, 171-182.
https://doi.org/10.1142/S1793557109000145
[7] Shahzad, N. (2004) Some General Random Coincidence Point Theorems. New Zealand Journal of Mathematics, 33, 95-103.
[8] Beg, I. and Shahzad, N. (1994) Random Approximations and Random Fixed Point Theorems. Journal of Applied Mathematics and Stochastic Analysis, 7, 145-150.
https://doi.org/10.1155/S1048953394000158
[9] Khan, A.R., Thaheem, A.B. and Hussain, N. (2002) Random Fixed Points and Random Approximations in Nonconvex Domains. Journal of Applied Mathematics and Stochastic Analysis, 15, 247-253.
https://doi.org/10.1155/S1048953302000217
[10] Itoh, S. (1979) Random Fixed Point Theorems with an Applications to Random Differential Equations in Banach Spaces. Journal of Mathematical Analysis and Applications, 67, 261-273.
https://doi.org/10.1016/0022-247X(79)90023-4
[11] Nashine, H.K. (2008) Random Fixed Points and Invariant Random Approximation in Non-Convex Domains. Hacettepe Journal of Mathematics and Statistics, 37, 81-88.
[12] Abed, S.S. and Ajeel, Y.J. (2016) On Invariant Random Best Approximation. International Journal of Recent Scientific Research, 7, 9530-9533.
[13] Ajeel, Y.J., Abed, S.S. and Alsaidy, S.K. (2016) On Random Fixed Points of Lipschitza in Random Operators. International Journal of Advanced Scientific and Technical Research, 5, 157-162.
[14] Alsaidy, S.K., Abed, S.S. and Ajeel, Y.J. (2015) On Random Coincidence Points and Random Best Approximation in p-Normed Spaces. Mathematical Theory and Modeling, 5, 41-48.
[15] Saluja, G.S. (2015) On Common Random Fixed Point Theorems under Contractive Type Condition in Cone Random Metric Spaces. Gulf Journal of Mathematics, 3, 111-122.
[16] Rashwan, R.A. and Albaqeri, D.M. (2012) A Common Random Fixed Point Theorem for Six Weakly Compatible Mappings in Hilbert Spaces. International Journal of Open Problems in Computer Science and Mathematics, 5, 34-46.
https://doi.org/10.12816/0006137
[17] Arunchai, A. and Plubtieng, S. (2013) Random Fixed Point Theorem of Krasnoselskii Type for the Sum of Two Operators. Journal of Fixed Point Theory and Applications, 142, 1-10.
https://doi.org/10.1186/1687-1812-2013-142
[18] Singh, B., Rathore, G.P.S., Dubey, P. and Singh, N. (2013) Common Unique Fixed Point Theorem for Random Operators in Hilbert Space. International Journal of Innovative Research in Science, 2, 6364-6370.
[19] Vishwakarma, N. and Chauhan, M.S. (2013) Common Random Fixed Point Theorems in Symmetric Spaces. International Journal of Advancements in Research and Technology, 2.
[20] Khanday, A.H., Jain, A. and Badshah, V.H. (2012) Random Fixed Point Theorems of Random Multivalued Operators on Polish Space. Ijrras, 12, 154-157.
[21] Chauhan, S.S. (2013) Common Fixed Point Theorem for Four Random Operators in Hilbert Space. Advances in Applied Science Research, 4, 219-222.
[22] Abed, S.S., Ajeel, Y.J. and Alsaidy, S.K. (2017) Results on Quasi Contraction Random Operators. International Journal of Science and Research (IJSR), 6, 607-610.
[23] Abed, S.S., Ajeel, Y.J. and Alsaidy, S.K. Results on General-Weakly Random Operators. To appear in International Journal of Pure and Applied Mathematics.

  
comments powered by Disqus

Copyright © 2019 by authors and Scientific Research Publishing Inc.

Creative Commons License

This work and the related PDF file are licensed under a Creative Commons Attribution 4.0 International License.