Next Article in Journal
Best Approximation Results in Various Frameworks
Previous Article in Journal
An Efficient Class of Traub–Steffensen-Type Methods for Computing Multiple Zeros
Previous Article in Special Issue
A Note on Anosov Homeomorphisms
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Generic Homeomorphisms with Shadowing of One-Dimensional Continua

by
Alfonso Artigue
1,* and
Gonzalo Cousillas
2,*
1
Departamento de Matemática y Estadística del Litoral, Universidad de la República, Salto 50000, Uruguay
2
Instituto de Matemática y Estadística “Rafael Laguardia”, Facultad de Ingeniería, Universidad de la República, Montevideo 11200, Uruguay
*
Authors to whom correspondence should be addressed.
Axioms 2019, 8(2), 66; https://doi.org/10.3390/axioms8020066
Submission received: 7 May 2019 / Revised: 23 May 2019 / Accepted: 24 May 2019 / Published: 26 May 2019
(This article belongs to the Special Issue Shadowing in Dynamical Systems)

Abstract

:
In this article, we show that there are homeomorphisms of plane continua whose conjugacy class is residual and have the shadowing property.
MSC:
37E05; 37C50; 37C20

1. Introduction

Let ( X , dist ) be a compact metric space and denote by H ( X ) the space of homeomorphisms f : X X with the C 0 distance
dist C 0 ( f , g ) = sup { dist ( f ( x ) , g ( x ) ) , dist ( f 1 ( x ) , g 1 ( x ) ) : x X } .
A property is said to be generic if it holds on a residual subset of H ( X ) . Recall that a set is G δ if it is a countable intersection of open sets and it is residual if it contains a dense G δ subset. For instance, it is known that the shadowing property is generic for X a compact manifold ([1], Theorem 1) or a Cantor set ([2], Theorem 4.3). Recall that f H ( X ) has the shadowing property if for all ε > 0 , there is δ > 0 such that if { x i } i Z is a δ -pseudo orbit, then there is y X such that dist ( f i ( y ) , x i ) < ε for all i Z . We say that { x i } i Z is a δ -pseudo orbit if dist ( f ( x i ) , x i + 1 ) < δ for all i Z .
A remarkable result, proved in [3,4], states that if X is a Cantor set, then there is a homeomorphism of X whose conjugacy class is a dense G δ subset of H ( X ) . That is, a generic homeomorphism of a Cantor set is conjugate to this special homeomorphism. We say that f , g H ( X ) are conjugate if there is h H ( X ) such that f h = h g and the conjugacy class of f is the set of all the homeomorphisms conjugate to f. This result gives rise to a natural question: besides the Cantor set,
which compact metric spaces have a Gδ dense conjugacy class?
On a space with a G δ dense conjugacy class, the study of generic properties (invariant under conjugacy, as the shadowing property) is reduced to determine whether a representative of this class has the property or not.
In Theorem 2, we show that there are one-dimensional plane continua with a G δ dense conjugacy class whose members have the shadowing property. The proof of this result is based on Theorem 1, where we show that for a compact interval I there is a G δ conjugacy class in H ( I ) which is dense in the subset of orientation preserving homeomorphisms of I. In addition, the proof of Theorem 2 depends on Propositions 2 and 3, where we give sufficient conditions for the existence of a residual conjugacy class and for a homeomorphism to have the shadowing property, respectively. The following open question has an affirmative answer in the examples known by the authors:
if a homeomorphism has a Gδ dense conjugacy class, does it have the shadowing property?

2. Generic Dynamics on a Closed Segment

Let I = [ 0 , 1 ] and define H + ( I ) = { f H ( I ) : f preserves orientation } . In this section, we show the following result.
Theorem 1.
There is f * H + ( I ) whose conjugacy class is a G δ dense subset of H + ( I ) .
Remark 1.
The generic dynamics of circle homeomorphisms is studied in detail in [5], Theorem 9.1. The proof of Theorem 1 follows the same ideas. As we could not find this result in the literature, we include the details.
To prove Theorem 1, we start by defining the homeomorphism f * . For this purpose, we introduce some definitions. For f H + ( I ) let fix ( f ) = { x X : f ( x ) = x } . A connected component of I \ fix ( f ) will be called a wandering interval. Following [6], we say that a wandering interval ( a , b ) is an r-interval if lim n + f n ( x ) = b for all x ( a , b ) . Analogously, it is an l-interval if lim n + f n ( x ) = a for all x ( a , b ) . For each interval [ a , b ] , fix a homeomorphism f r [ a , b ] : [ a , b ] [ a , b ] such that ( a , b ) is an r-interval. Analogously, we consider f l [ a , b ] with ( a , b ) an l-interval.
For n 0 and 0 k < 3 n , define the closed interval
J ( n , k ) = 3 k + 1 3 n + 1 , 3 k + 2 3 n + 1 .
For x in the ternary Cantor set, define f * ( x ) = x . In another case, there is a minimum integer n x 0 such that x J ( n x , k ) for some 0 k < 3 n x and define
f * ( x ) = f l J ( n x , k ) ( x ) if n x is odd , f r J ( n x , k ) ( x ) if n x is even .
For example, ( 1 3 , 2 3 ) is an r-interval, while ( 1 3 2 , 2 3 2 ) and ( 7 3 2 , 8 3 2 ) are l-intervals. See Figure 1.
Remark 2.
From [7], Theorem 8, we know that f * , and every homeomorphism conjugate to f * , has the shadowing property.
The next result gives a useful characterization of the conjugacy class of f * . Given ε > 0 , we say that g H + ( I ) satisfies the property P ε if there are intervals J i = ( a i , b i ) , i = 1 , , n , such that:
  • 0 < a 1 < b 1 < a 2 < b 2 < a 3 < < b n < 1 ;
  • J i is an r-interval for i odd and an l-interval for i even;
  • max { a 1 , 1 b n } < ε and max { a i + 1 b i : 1 i < n } < ε .
Proposition 1.
A homeomorphism g H + ( I ) is conjugate to f * if and only if it satisfies P ε for all ε > 0 .
Proof. 
The direct part of the proof is clear from the construction of f * .
To prove the converse, suppose that g satisfies P ε for all ε > 0 . From Condition (3), we see that fix ( g ) is totally disconnected. Suppose that p I is an isolated fixed point. If p = 0 , then there is a wandering interval ( 0 , x ) . Taking ε ( 0 , x ) , we have a contradiction with (3), because a 1 < ε . Analogously we show that p cannot be 1. If p ( 0 , 1 ) , then p is in the boundary of two wandering intervals. Taking ε smaller than the length of these intervals, we contradict (1) and (3). Thus, fix ( g ) has no isolated point and is a Cantor set. Condition (2) (applied for a suitable ε small) implies that between two wandering intervals there is an r-interval and an l-interval.
Let R and L be the families of r-intervals and l-intervals of g, respectively. We define an order in R L in the following way: I α < I β if x < y for all x I α , y I β . We will make the conjugacy by induction. For the first step, name I 1 / 2 R which satisfies diam ( I 1 / 2 ) diam ( I ) for every I R . In the case that there exists more than one interval which verifies this condition, we choose any of them. Let J c be a wandering interval of f * such that c is the midpoint of J c . By construction, J 1 / 2 is an r-interval of f * , thus we can consider a conjugacy h 1 / 2 : I 1 / 2 J 1 / 2 of g and f * restricted to these intervals. Notice that 1 / 6 and 5 / 6 are the midpoints of ( 1 / 9 , 2 / 9 ) and ( 7 / 9 , 8 / 9 ) , respectively. Take I 1 / 6 L satisfying I 1 / 6 < I 1 / 2 and diam ( I 1 / 6 ) diam ( I ) for every I L such that I < I 1 / 2 . In addition, take I 5 / 6 L satisfying I 1 / 2 < I 5 / 6 and diam ( I 5 / 6 ) diam ( I ) for every I L such that I > I 1 / 2 . Then, consider h 1 / 6 : I 1 / 6 J 1 / 6 to be a conjugacy from g to f * restricted to the corresponding intervals. Similarly, define h 5 / 6 . Then, we go on defining 2 k 1 homeomorphisms on each step. If k 1 is even, we choose r-intervals, otherwise we choose l-intervals. Notice that since in each step we choose the largest interval of the r or l-intervals of g, every wandering interval of g is eventually chosen. In this way, the conjugacies h j / k give rise to a conjugacy h of g and f * in the whole interval [ 0 , 1 ] and the proof ends. □
Proof of Theorem 1.
Given n 1 , let U n be the set of increasing homeomorphisms of I satisfying P 1 / n . Notice that P ε implies P ε for all ε > ε > 0 . Thus, from Proposition 1 we have that the conjugacy class of f * is the countable intersection n 1 U n . To finish the proof, applying Baire’s Theorem, we show that each U n is open and dense in H + ( I ) .
To prove that U n is open, consider f U n . It is clear that there is δ > 0 such that f U n 4 δ . Consider the intervals ( a i , b i ) from the definition of property P ε , for ε = 1 / n . For each odd i = 1 , , n , take x i ( a i , a i + δ ) and for i even take y i ( b i δ , b i ) . Consider m N large such that f m ( x i ) ( b i δ , b i ) and f m ( y i ) ( a i , a i + δ ) for all i. Take a neighborhood V of f such that dist C 0 ( f m , g m ) < δ for all g V and g m ( x i ) > x i , g m ( y i ) < y i for all i. This implies that ( x i , g m ( x i ) ) is contained in an r-interval for g and ( g m ( y i ) , y i ) is contained in an l-interval for g. For all g V and i odd, we have
| g m ( x i ) g m ( y i + 1 ) ) | | g m ( x i ) f m ( x i ) | + | f m ( x i ) f m ( y i + 1 ) | + | f m ( y i + 1 ) g m ( y i + 1 ) | < δ + | f m ( x i ) b i | + | b i a i + 1 | + | a i + 1 f m ( y i + 1 ) | + δ < 2 δ + ( 1 / n 4 δ ) + 2 δ = 1 / n .
Arguing analogously for i even, we conclude that g U n and U n is open.
To prove that U n is dense in H + ( I ) , the following remark is sufficient. Given f H + ( I ) , p fix ( f ) ( 0 , 1 ) and δ > 0 small, we can define g H + ( I ) close to f such that:
  • f | [ 0 , p ] and g | [ 0 , p δ ] are conjugate;
  • f | [ p , 1 ] and g | [ p + δ , 1 ] are conjugate; and
  • g has an r or l-interval at [ p δ , p + δ ] .
That is, a fixed point can be exploded into a small wandering interval with an arbitrarily small perturbation. By finitely performing many such explosions, the density of U n is obtained. □

3. Genericity on a Plane One-Dimensional Continuum

In this section, we show that there are some particular one-dimensional plane continua with a G δ dense conjugacy class whose members have the shadowing property. We start with a sufficient condition for the existence of a G δ dense conjugacy class. An open subset U X is a free arc if it is homeomorphic to R .
Proposition 2.
If X is a compact metric space such that
  • X = n 1 a n , where each a n is a compact arc with extreme points p n , q n X for all n 1 ;
  • a n \ { p n , q n } is a free arc for all n 1 ; and
  • for all f H ( X ) , it holds that f ( a n ) = a n and p n , q n fix ( f ) for all n 1 ;
then H ( X ) has a G δ dense conjugacy class.
Proof. 
For each n 1 , let X n = clos ( X \ a n ) and define
H n = { f H ( X n ) : p n , q n fix ( f ) } ,
and the map φ n : H ( X ) H + ( a n ) × H n as φ n ( f ) = f | a n × f | X n . In H + ( a n ) × H n , we consider the product topology. It is clear that φ n is a homeomorphism for each n 1 . Let R n be the G δ dense conjugacy class of H + ( a n ) given by Theorem 1 and define S n = R n × H n . Thus, n 1 φ n 1 ( S n ) is a G δ dense conjugacy class in H ( X ) . □
Remark 3.
Notice that a representative g * of the G δ dense conjugacy of Proposition 2 is obtained by considering a conjugate of f * on each arc a n of X.
Now, we prove a sufficient condition for a homeomorphism to have the shadowing property. For this purpose, we need some definitions and a lemma. Suppose that ( X , dist ) is a compact metric space and take f H ( X ) . A compact f-invariant subset A X is a quasi-attractor if for every open neighborhood U of A there is an open subset V U such that A V and clos ( f ( V ) ) V . If, in addition, f : A A has the shadowing property, we say that A is a quasi-attractor with shadowing.
Lemma 1.
If A X is a quasi-attractor with shadowing, then for all ε > 0 there is δ > 0 such that if { x n } n 0 is a δ-pseudo-orbit with x 0 B δ ( A ) , then there is y A that ε-shadows { x n } n 0 .
Proof. 
Given ε > 0 , take δ 1 > 0 such that every δ 1 -pseudo-orbit in A is ε / 2 -shadowed by a point in A. Consider 0 < α < min { ε / 2 , δ 1 / 3 } such that dist ( a , b ) < α implies dist ( f ( a ) , f ( b ) ) < δ 1 / 3 . Since A is a quasi-attractor, for U = B α ( A ) there exists an open set V such that A V U and clos ( f ( V ) ) V . Take δ ( 0 , δ 1 / 3 ) such that B δ ( clos ( f ( V ) ) ) V .
Suppose that { x n } n 0 is a δ -pseudo-orbit with x 0 B δ ( A ) . Since f ( x 0 ) f ( V ) , we have that x 1 B δ ( f ( V ) ) and x 1 V . In this way, we prove that x n V for all n 0 . For each n 0 , take y n A such that dist ( y n , x n ) < α . We have that
dist ( f ( y n ) , y n + 1 ) dist ( f ( y n ) , f ( x n ) ) + dist ( f ( x n ) , x n + 1 ) + dist ( x n + 1 , y n + 1 ) δ 1 / 3 + δ + α < 3 δ 1 / 3 = δ 1 .
This proves that { y n } n 0 is a δ 1 -pseudo-orbit contained in A. There exists z A that ε / 2 -shadows { y n } n 0 . Thus,
dist ( f n ( z ) , x n ) dist ( f n ( z ) , y n ) + dist ( y n , x n ) < ε / 2 + α ε .
Therefore, the proof ends. □
Proposition 3.
If every point of X belongs to a quasi-attractor with shadowing, then f has shadowing.
Proof. 
Suppose that ε > 0 is given. For each x X , let A x X be a quasi-attractor with shadowing containing x. Let δ x > 0 be given by Lemma 1 such that for every δ x -pseudo-orbit { x n } n 0 with x 0 B δ x ( A x ) there is a point in A x that ε -shadows { x n } n 0 . As X is compact, there is a finite sequence x 1 , , x k X such that i = 1 k B δ i ( A i ) = X , where A i = A x i and δ i = δ x i . If we take δ = min { δ 1 , , δ k } , we have that for every δ -pseudo-orbit { x n } n 0 in X, there is j such that x 0 B δ j ( A j ) . Then, there is a point in A j that ε -shadows { x n } n 0 and the proof ends. □
Let Y R 2 be the union of
  • the circle arc x 2 + y 2 = 1 , y 0 ;
  • the horizontal segment [ 1 , 1 ] × { 0 } ; and
  • the vertical segments { 1 + 2 n } × [ 0 , 1 / n ] , for n 1 .
Theorem 2.
For the continuum Y, there is a G δ conjugacy class which is dense in H ( Y ) and whose members have the shadowing property. In particular, the shadowing property is generic in H ( Y ) .
Proof. 
The continuum Y satisfies the hypothesis of Proposition 2. Indeed, the conditions (1) and (2) are directly from the construction of Y. Consider the points p n , p ˜ indicated in Figure 2. It is clear that p ˜ fix ( f ) for all f H ( Y ) . This implies that a 1 is invariant and p 2 fix ( f ) . In turn, this implies that a 2 is invariant under each f H ( Y ) . In this way, it is shown that condition (3) of Proposition 2 holds. Therefore, H ( Y ) contains a G δ dense conjugacy class.
As explained in Remark 3, a representative g * H ( Y ) of this conjugacy class is obtained by taking a conjugate of f * on each arc a n . It only remains to prove that g * has the shadowing property. By Remark 2, we know that g * : a n a n has the shadowing property. By construction, each a n is a quasi-attractor for g * . Since the arcs a n cover Y, we can apply Proposition 3 to conclude that g * has the shadowing property. □

Author Contributions

Both authors contributed equally to this work.

Funding

This research received no external funding.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Pilyugin, S.Y.; Plamenevskaya, O.B. Shadowing is generic. Topol. Appl. 1999, 97, 253–266. [Google Scholar] [CrossRef] [Green Version]
  2. Bernardes, N.C., Jr.; Darji, U.B. Graph theoretic structure of maps of the Cantor space. Adv. Math. 2012, 231, 1655–1680. [Google Scholar] [CrossRef] [Green Version]
  3. Akin, E.; Glasner, E.; Weiss, B. Generically there is but one self homeomorphism of the Cantor set. Trans. Am. Math. Soc. 2008, 360, 3613–3630. [Google Scholar] [CrossRef]
  4. Kechris, A.S.; Rosendal, C. Turbulence, amalgamation, and generic automorphisms of homogeneous structures. Proc. Lond. Math. Soc. 2007, 94, 302–350. [Google Scholar] [CrossRef]
  5. Akin, E.; Hurley, M.; Kennedy, J.A. Dynamics of Topologically Generic Homeomorphisms; American Mathematical Society: Providence, RI, USA, 2003. [Google Scholar]
  6. Pilyugin, S.Y. Shadowing in Dynamical Systems; Lecture Notes in Mathematics; Springer: Berlin, Germany, 1999; Volume 1706. [Google Scholar]
  7. Pennings, T.; Eeuwen, J.V. Pseudo-orbit shadowing on the unit interval. Real Anal. Exchang. 1990, 16, 238–244. [Google Scholar]
Figure 1. A sketch of the phase diagram of f * .
Figure 1. A sketch of the phase diagram of f * .
Axioms 08 00066 g001
Figure 2. The continuum Y can be decomposed as a union of arcs as in Proposition 2.
Figure 2. The continuum Y can be decomposed as a union of arcs as in Proposition 2.
Axioms 08 00066 g002

Share and Cite

MDPI and ACS Style

Artigue, A.; Cousillas, G. Generic Homeomorphisms with Shadowing of One-Dimensional Continua. Axioms 2019, 8, 66. https://doi.org/10.3390/axioms8020066

AMA Style

Artigue A, Cousillas G. Generic Homeomorphisms with Shadowing of One-Dimensional Continua. Axioms. 2019; 8(2):66. https://doi.org/10.3390/axioms8020066

Chicago/Turabian Style

Artigue, Alfonso, and Gonzalo Cousillas. 2019. "Generic Homeomorphisms with Shadowing of One-Dimensional Continua" Axioms 8, no. 2: 66. https://doi.org/10.3390/axioms8020066

APA Style

Artigue, A., & Cousillas, G. (2019). Generic Homeomorphisms with Shadowing of One-Dimensional Continua. Axioms, 8(2), 66. https://doi.org/10.3390/axioms8020066

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