Strict fixed point problem, stability results and retraction displacement condition for Picard operators

Cristina Gheorghe, Adrian Petruşel · 2025

Read Strict fixed point problem, stability results and retraction displacement condition for Picard operators by Cristina Gheorghe, Adrian Petruşel (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-00207.txt.

Introduction

Cristina Gheorghe and Adrian Petruşel wrote this math paper in 2025 on fixed points of set-valued maps. A strict fixed point is a point the map sends to itself alone. The pair ask when such a point exists, stands alone, and draws all repeat uses of the map toward it. Their maps obey shrinking rules of two classic kinds, tied to the names Ćirić and Reich-Rus.

They prove rules that guarantee the lone strict point. They then check how the finding holds up when the map is nudged, a trait called data dependence. Other steadiness tests follow: one named for Ulam and Hyers, one for well-posedness in the sense of Reich and Zaslavski, and the Ostrowski trait. A retraction-based displacement rule frames the work. The paper serves analysts who need maps with many outputs to still land on one sure point.

Catalog record

CREATORCristina Gheorghe, Adrian Petruşel
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE26 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-00207

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Strict fixed point problem, stability results and retraction displacement condition for Picard operators Cristina Gheorghe, Adrian Petruşel Abstract The aim of this paper is to give strict fixed point principles for multivalued operators T : X → P ⁡ ( X ) T:X\rightarrow P(X) satisfying some contraction conditions of Ćirić and of Ćirić-Reich-Rus type. We are interested, under which conditions, the multi-valued operator has a unique strict fixed point and, additionally, when the sequence of its multi-valued iterates ( T n ​ ( x ) ) n ∈ ℕ (T^{n}(x))_{n\in\mathbb{N}} converges to this unique strict fixed point. Moreover, some stability properties, such as data dependence on operator perturbation, Ulam-Hyers stability, well-posedness in the sense of Reich and Zaslavski and Ostrowski property of the strict fixed point problem are established. 2010 Mathematics Subject Classification]47H10, 54H25 keywords strict fixed point, multi-valued operator, data dependence, Ulam-Hyers stability, well-posedness, Ostrowski property, complete metric space, Picard operators, contraction-type conditions pacs [ 1 Introduction The main metric fixed point theorem for multivalued contractions on complete metric spaces was given by Nadler in 1969. Then, one year later, Covitz and Nadler extended the result for the case of a generalized metric space. Later on, there were obtained many other extensions of these results. For example, A. Petruşel and G. Petruşel gave in [ 5 ] a saturated fixed point principle for multivalued contractions. The purpose of this paper is to formulate similar principles for some elements of the class multivalued Picard operators. More precisely, Ćirić type operators (see [ 9 ] ) and Ćirić-Reich-Rus type operators (see [ 2 ] ) will be considered. Our goal is to determine under which assumptions the multi-valued operator has a unique strict fixed point and, moreover, when the sequence of its multivalued iterates (T^{n}(x))_{n\in\mathbb{N}} is convergent to this unique fixed point. We also study some stability properties, such as data dependence problem, Ulam-Hyers stability, well-posedness in the sense of Reich and Zaslavski and Ostrowski property for the strict fixed point problem. Terminology and notations We will use the consecrated notions (see, also [ 8 ] , [ 11 ] ). In this paper, for a metric space (X,d) , P(X) stands for the family of all non-empty subsets of X, P_{cp} represent the family of all non-empty compact subsets of X, and P_{cl} is the family of all non-empty closed subsets of X. Also, for a multivalued operator T:X\rightarrow P(X) , the following notations are used: • the fixed point set of T: \displaystyle Fix(T):=\{x\in X:x\in T(x)\} • the strict fixed point set of T: \displaystyle SFix(T):=\{x\in X:\{x\}=T(x)\} • the graphic of T: \displaystyle Graph(T):=\{(x,y)\in X\times X:y\in T(x)\} We also recall, in the context of a metric space, the definitions of the following functionals: • the gap functional generated by d : \displaystyle D:P(X)\times P(X)\rightarrow\mathbb{R}_{+},~D(A,B):=\inf\{d(a,b):a\in A,b\in B\} • the excess functional of A over B generated by d: \displaystyle e:P(X)\times P(X)\rightarrow\mathbb{R}_{+}\cup\{+\infty\},~e(A,B):=\sup\{D(a,B):a\in A\}; • the Pompeiu-Hausdorff functional generated by d : \displaystyle H:P(X)\times P(X)\rightarrow\mathbb{R}_{+}\cup\{+\infty\},~H(A,B):=\max\{e(A,B),e(B,A)\}. Definition 1.1 . Let (X,d) be a metric space. The operator T:X\rightarrow P(X) is a multivalued weakly Picard operator (MWP operator) if, for each x\in X and for each y\in T(x) , there exists a sequence (x_{n})_{n\in\mathbb{N}} in X which satisfies the following conditions: i) x_{0}=x,x_{1}=y ; ii) x_{n+1}\in T(x_{n}) , for all n\in\mathbb{N} ; iii) the sequence (x_{n})_{n\in\mathbb{N}} is convergent and its limit is a fixed point of T. Remark 1.2 . Any sequence satisfying conditions i) and ii) from Definition 1.1 , is called a sequence of successive approximations of T. Definition 1.3 . We call T:X\rightarrow P(X) a multivalued Picard (MP) operator if the following statements hold true: i) SFix(T)=Fix(T)=\{x^{*}\} ; ii) T^{n}(x)\xrightarrow{H}\{x^{*}\} as n\rightarrow\infty , for each x\in X. Definition 1.4 . A multivalued Picard operator T:X\rightarrow P(X) is called \Psi- multivalued Picard operator ( \Psi-MP operator) if there exists an increasing mapping \Psi:\mathbb{R_{+}}\rightarrow\mathbb{R_{+}} with \Psi continuous at 0 and \Psi(0)=0 , such that d(x,x^{*})\leq\Psi(D(x,T(x)),\text{ for all }x\in X. Remark 1.5 . The above condition is also called the strong retraction-displacement condition for the case of unique (strict) fixed point of T . 2 Main results for multivalued Ćirić type contractions Retraction-displacement condition In this section, we will give two main fixed point principles for multivalued retractions satisfying some contraction-type conditions. In a similar manner as in Theorem 5.5 from [ 5 ] , we can formulate the following first main result: Theorem 2.1 . Let (X,d) be a complete metric space and T:X\rightarrow P_{cl}(X) be an operator such that SFix(T)\neq\varnothing . Suppose there exist \alpha,\beta,\gamma\geq 0 with \alpha+\beta+\gamma<1 such that \displaystyle H(T(x),T(y)) \displaystyle\leq\alpha d(x,y)+\beta D(x,T(y))+\gamma D(y,T(x)),\text{ for all }x,y\in X. Then i) Fix(T)=SFix(T)=\{x^{*}\} ; ii) for all x\in X , the sequence of sets (T^{n}(x))_{n\in\mathbb{N}} converges to \{x^{*}\} with respect to the Pompeiu-Hausdorff metric H ; iii) d(x,x^{*})\leq\dfrac{1+\gamma}{(1-\alpha-\beta)\xi}D(x,T(x)) , for all x\in X , where \xi\in(0,1) can be arbitrary chosen. Proof. i) Let x^{*}\in SFix(T) . Suppose there exists y^{*}\in SFix(T) with x^{*}\neq y^{*} . Then, we have: \displaystyle d(x^{*},y^{*}) \displaystyle=H(T(x^{*}),T(y^{*})) \displaystyle\leq\alpha d(x^{*},y^{*})+\beta D(y^{*},T(x^{*}))+\gamma D(x^{*},T(y^{*})) \displaystyle=\alpha d(x^{*},y^{*})+\beta d(x^{*},y^{*})+\gamma d(x^{*},y^{*}). Therefore, (1-\alpha-\beta-\gamma)d(x^{*},y^{*})\leq 0. Since \alpha+\beta+\gamma<1 , it follows that d(x^{*},y^{*})=0 , so x^{*}=y^{*} . The contradiction proves that SFix(T)=\{x^{*}\} . Let \tilde{x}\in Fix(T) with x^{*}\neq\tilde{x} . Then, we have \displaystyle d(\tilde{x},x^{*}) \displaystyle=D(\tilde{x},T(x^{*}))\leq H(T(\tilde{x}),T(x^{*})) \displaystyle\leq\alpha d(\tilde{x},x^{*})+\beta D(\tilde{x},T(x^{*}))+\gamma D(x^{*},T(\tilde{x})) \displaystyle\leq\alpha d(x^{*},\tilde{x})+\beta d(\tilde{x},x^{*})+\gamma d(x^{*},\tilde{x}). We get again a contradiction and thus Fix(T)=\{x^{*}\} . ii) Let x\in X be an arbitrary element and consider the sequence (T^{n}(x))_{n\in\mathbb{N}} . Then, we have: \displaystyle H(T(x),x^{*})=H(T(x),T(x^{*})) \displaystyle\leq\alpha d(x,x^{*})+\beta D(x,T(x^{*}))+\gamma D(x^{*},T(x)) \displaystyle\leq\alpha d(x,x^{*})+\beta d(x,x^{*})+\gamma H(T(x),x^{*}). Thus, with the notation k:=\dfrac{\alpha+\beta}{1-\gamma}\in(0,1) , we get: H(T(x),x^{*})\leq kd(x,x^{*}). We prove, by mathematical induction, that the following proposition is true P(n):H(T^{n}(x),\{x^{*}\})\leq k^{n}d(x,x^{*}),\text{ for all }n\in\mathbb{N}^{*}. By the above proof, it is clear that P(1) is true. We will prove that P(n)\Rightarrow P(n+1). We show first that H(T(Y),x^{*})\leq kH(Y,x^{*}) , for all Y\in P_{cl}(X). Indeed, let v\in T(Y) be an arbitrary element. Then, there exists u\in Y such that v\in T(u) . We have d(v,x^{*})\leq H(T(u),\{x^{*}\})\leq kd(u,x^{*})\leq kH(Y,\{x^{*}\}). Taking \displaystyle\sup_{v\in T(Y)} from the previous inequality, we get that H(T(Y),\{x^{*}\})\leq kH(Y,\{x^{*}\}). (2.1) Therefore, we obtain: \displaystyle H(T^{n+1}(x),\{x^{*}\}) \displaystyle=H(T(T^{n}(x)),\{x^{*}\}) \displaystyle\overset{(\ref{eq})}{\leq}kH(T^{n}(x),\{x^{*}\}) \displaystyle\overset{P(n)}{\leq}k^{n+1}H(T(x),\{x^{*}\}). The proof of the mathematical induction is now complete. Hence, H(T^{n}(x),\{x^{*}\})\leq k^{n}d(x,x^{*})\rightarrow 0\text{ as }n\rightarrow\infty. Therefore, T^{n}(x)\overset{H}{\rightarrow}\{x^{*}\} as n\rightarrow\infty. iii) Let \xi\in(0,1) be arbitrary and let x\in X . Then, there exists \widetilde{y}\in T(x) such that \xi d(x,\widetilde{y})\leq D(x,T(x)) , or, equivalently, d(x,\widetilde{y})\leq\dfrac{1}{\xi}D(x,T(x)). (2.2) On the other hand, \displaystyle d(x,x^{*}) \displaystyle\leq d(x,\widetilde{y})+d(\widetilde{y},x^{*}) \displaystyle=d(x,\widetilde{y})+D(\widetilde{y},T(x^{*})) \displaystyle\leq d(x,\widetilde{y})+H(T(x),T(x^{*})) \displaystyle\leq d(x,\widetilde{y})+\alpha d(x,x^{*})+\beta D(x,T(x^{*}))+\gamma D(x^{*},T(x)) \displaystyle\leq d(x,\widetilde{y})+\alpha d(x,x^{*})+\beta d(x,x^{*})+\gamma d(x^{*},\widetilde{y}). Hence, (1-\alpha-\beta)d(x,x^{*})\leq(1+\gamma)d(x,\widetilde{y}). (2.3) From ( 2.3 ) and ( 2.2 ), we get the relation iii). ∎ For further results related to retractions of a fixed point set, we refer to [ 5 ] , [ 13 ] , [ 6 ] , [ 12 ] , [ 14 ] . Well-posedness of the strict fixed point problem Definition 2.2 . The strict fixed point problem T(x)=\{x\} is said to be well-posed in the sense of Reich and Zaslavski (see [ 9 ] , [ 10 ] ) if SFix(T)=\{x^{*}\} and the following implication holds: (u_{n})_{n\in\mathbb{N}}\subset X\text{ and }D(u_{n},T(u_{n}))\rightarrow 0\implies u_{n}\rightarrow x^{*} as n\rightarrow\infty . Theorem 2.3 . Let (X,d) be a complete metric space and T:X\rightarrow P_{cl}(X) be an operator such

Read the complete text