Preprint
Article

This version is not peer-reviewed.

Devaney Chaos for the Function Envelope of a Non-Autonomous Discrete Dynamical System

Submitted:

01 August 2026

Posted:

04 August 2026

You are already at the latest version

Abstract
This paper investigates Devaney chaos in the \(L^{1}\)-function envelope system associated with a nonautonomous discrete dynamical system. We prove that if (I, f) is weakly mixing of order 3, then its \(L^1\)-functional envelope system is weakly mixing of all orders. Moreover, if, in addition, the periodic points of (I, f) are dense in \(I\), then the functional envelope system \((L^1(I,I),H_{\infty})\) is Devaney chaotic.
Keywords: 
;  ;  

1. Introduction

A nonautonomous dynamical system, abbreviated as an NDS, is a pair ( X , f ) , where X is a compact metric space endowed with a metric d, and f = { f n } n = 1 is a sequence of continuous self-maps on X. Nonautonomous dynamical systems were systematically studied in the framework of topological dynamics by Kolyada and Snoha [1]. For integers 1 m n , put f m , n = f n f n 1 f m , and set f m , m 1 = id X . Then the positive orbit of x 1 is defined by
x n + 1 = f 1 , n ( x 1 ) , n N + .
If f n = f for every n 1 , then ( X , f ) reduces to the classical autonomous dynamical system ( X , f ) .
A dynamical system ( X , f ) is said to be Devaney chaotic [2] if it satisfies the following conditions:
1.
( X , f ) is topologically transitive;
2.
the set of periodic points of f is dense in X;
3.
( X , f ) has sensitive dependence on initial conditions.
Banks et al. [3] proved that, in the definition of Devaney chaos, the sensitivity condition is redundant.
Functional envelope systems provide a natural framework for studying the relationship between the dynamics on an underlying space and the dynamics on an infinite-dimensional function space. A classical discrete dynamical system describes the evolution of a single initial point under successive iterations of continuous mappings. In contrast, a functional envelope system takes functions as its states and describes, through composition operators, the simultaneous evolution of a family of initial states, spatial distributions, or global state profiles.
More precisely, for every n N + , define H n ( φ ) = f n φ . The sequence H = { H n } n = 1 generates the L 1 -functional envelope system. Its n-step cumulative map is given by H 1 , n ( φ ) = f 1 , n φ . Here, a function φ L 1 ( I , I ) no longer represents a single state in the underlying space. Instead, it may be regarded as a family of states parametrized by points of I. Consequently, the functional envelope system describes the collective evolution of a large family of initial states under the same nonautonomous dynamical process.
The functional envelope construction lifts a dynamical problem on a finite- or low-dimensional underlying space to an infinite-dimensional function space. This makes it possible to investigate whether dynamical properties such as transitivity, weak mixing, sensitivity, and periodic structure are preserved under this lifting procedure, and whether the chaotic behavior of the functional envelope system can, conversely, characterize the dynamics of the underlying system.
In the autonomous setting, Chen et al. [4] proved that weak mixing of an autonomous interval system implies Devaney chaos of its L 1 -functional envelope system. Several other dynamical properties of functional envelope systems associated with autonomous systems have also been studied; see, for example, [5,6,7,8,9]. For nonautonomous discrete dynamical systems, Chen et al. [10] mainly investigated transitivity, sensitivity, and Li–Yorke chaos of the corresponding functional envelope systems.
It is therefore natural and important to determine whether density of periodic points can be transferred between an underlying nonautonomous system and its functional envelope system. In particular, Sánchez et al. [11] showed that, for a nonautonomous discrete dynamical system, density of periodic points in the original system does not necessarily imply density of periodic points in the corresponding hyperspace system. Thus, identifying a mechanism that preserves the density of periodic points in a functional envelope system is a crucial step toward establishing a theory of Devaney chaos on function spaces. There also exist several research results concerning non-autonomous dynamical systems [13,14,15,16].
In this paper, we prove that density of periodic points in a nonautonomous interval system is equivalent to density of periodic points in its L 1 -functional envelope system. Combining this equivalence with a lifting theorem for weak mixing of all orders, we further establish Devaney chaos of the functional envelope system.
Our main result is stated as follows.
Theorem 1. 
Let ( I , f ) be a nonautonomous dynamical system on the interval I = [ 0 , 1 ] . Suppose that
1.
( I , f ) is weakly mixing of order 3;
2.
the set of periodic points of ( I , f ) is dense in I.
Then the functional envelope system ( L 1 ( I , I ) , H ) is Devaney chaotic.

2. Preliminaries

Throughout this paper, N + and N represent positive integer sets and non-negative integer sets, respectively, and let I = [ 0 , 1 ] .
A nonautonomous discrete dynamical system on I is denoted by ( I , f ) , f = { f n } n = 1 , where each map f n : I I is continuous. For integers 1 m n , define f m , n = f n f n 1 f m . In particular, f 1 , n = f n f n 1 f 1 . We also put f 1 , 0 = id I .
Let L 1 ( I , I ) = g : I I : g is Lebesgue measurable . Define an equivalence relation ∼ on L 1 ( I , I ) by
φ ψ I | φ ( t ) ψ ( t ) | d μ ( t ) = 0 .
The quotient space is denoted by L 1 ( I , I ) = L 1 ( I , I ) / . It is equipped with the metric
ρ ( φ , ψ ) = I | φ ( t ) ψ ( t ) | d μ ( t ) , φ , ψ L 1 ( I , I ) ,
where μ denotes the Lebesgue measure on I.
For every n N + , define the one-step functional envelope map H n : L 1 ( I , I ) L 1 ( I , I ) by H n ( φ ) = f n φ . The sequence H = { H n } n = 1 generates the L 1 -functional envelope system L 1 ( I , I ) , H .
For integers 1 m n , define the cumulative map H m , n = H n H n 1 H m . We also put H 1 , 0 = id L 1 ( I , I ) . Then, for every φ L 1 ( I , I ) , H m , n ( φ ) = f m , n φ . In particular,
H 1 , n ( φ ) = H n H n 1 H 1 ( φ ) = f n f n 1 f 1 φ = f 1 , n φ .
It is readily verified that L 1 ( I , I ) , ρ is a separable complete metric space.
A point a I is called a periodic point of ( I , f ) if there exists q N + such that
f 1 , m + q ( a ) = f 1 , m ( a ) for every m N 0 .
The least positive integer q satisfying the above equality is called the period of a. The set of all periodic points of ( I , f ) is denoted by Per ( f ) .
The system ( I , f ) is said to be topologically transitive if, for any two nonempty open subsets U , V I , there exists n N + such that f 1 , n ( U ) V .
For nonempty open subsets U , V I , define the hitting-time set by
N f ( U , V ) = n N + : f 1 , n ( U ) V .
The system ( I , f ) is said to be weakly mixing of order m if, for any nonempty open subsets U 1 , , U m , V 1 , , V m I , one has
i = 1 m N f ( U i , V i ) .
It is said to be weakly mixing of all orders if it is weakly mixing of order m for every integer m 2 .
The system ( I , f ) is said to have sensitive dependence on initial conditions if there exists η > 0 such that, for every x I and every ε > 0 , there exist y B I ( x , ε ) and n N + such that
f 1 , n ( x ) f 1 , n ( y ) > η .
We now present several key properties and lemmas.
Lemma 1. 
For every n N + , the one-step map H n : L 1 ( I , I ) L 1 ( I , I ) is well defined and continuous. Consequently, for every n N + , the cumulative map H 1 , n = H n H n 1 H 1 is continuous.
Proof of Lemma 1. 
Fix n N + . We first prove that H n is well defined. Let φ L 1 ( I , I ) and choose a measurable representative of its equivalence class. Since f n is continuous, the composition f n φ is measurable. Moreover,
0 f n ( φ ( t ) ) 1
for almost every t I , and hence f n φ is Lebesgue integrable.
Suppose that φ = ψ in L 1 ( I , I ) . Then φ ( t ) = ψ ( t ) for almost every t I . Consequently, f n ( φ ( t ) ) = f n ( ψ ( t ) ) for almost every t I . Thus, f n φ and f n ψ determine the same equivalence class. Therefore, H n is well defined.
We next prove that H n is continuous. Let ε > 0 . Since f n is continuous on the compact interval I, it is uniformly continuous. Hence there exists δ > 0 such that
| s t | < δ | f n ( s ) f n ( t ) | < ε 2
for all s , t I .
Set η = δ ε 2 . Suppose that ρ ( φ , ψ ) < η . Define
A = { t I : | φ ( t ) ψ ( t ) | < δ } , B = I A .
By Chebyshev’s inequality,
μ ( B ) 1 δ I | φ ( t ) ψ ( t ) | d μ ( t ) = ρ ( φ , ψ ) δ < ε 2 .
For A, we have
| f n ( φ ( t ) ) f n ( ψ ( t ) ) | < ε 2 .
Since f n ( I ) I = [ 0 , 1 ] , for B we have
| f n ( φ ( t ) ) f n ( ψ ( t ) ) | 1 .
Therefore,
ρ H n ( φ ) , H n ( ψ ) = I | f n ( φ ( t ) ) f n ( ψ ( t ) ) | d μ ( t ) A ε 2 d μ ( t ) + B 1 d μ ( t ) ε 2 μ ( A ) + μ ( B ) < ε .
Thus H n is continuous.
Finally, H 1 , n = H n H n 1 H 1 is a finite composition of continuous maps, and hence it is continuous. □
Lemma 2. 
[12] Theorem 11. Suppose that ( I , f ) is weakly mixing of order 3. Then it is weakly mixing of order n for every n 2 .

3. Proof of the Main Results

Theorem 2. 
Let ( I , f ) be a nonautonomous discrete dynamical system on I = [ 0 , 1 ] , and let L 1 ( I , I ) , H be its L 1 -functional envelope system. Then the following statements are equivalent:
1. 
Per ( f ) is dense in I;
2. 
Per ( H ) is dense in L 1 ( I , I ) .
Proof of Theorem 2. 
(1) ⟹ (2). Suppose that the set of periodic points of ( I , f ) is dense in I. Let g L 1 ( I , I ) and ε > 0 be arbitrary.
Since C ( I , I ) is dense in L 1 ( I , I ) , there exists h C ( I , I ) such that ρ ( g , h ) < ε 2 . Choose n N + sufficiently large so that 1 n < ε 2 . Define the measurable subsets
J i = t I : i 1 n h ( t ) < i n , i = 1 , , n 1 , J n = t I : n 1 n h ( t ) 1 .
Then J 1 , , J n form a measurable partition of I.
By the density of the periodic points of ( I , f ) , for each i = 1 , , n , we may choose a periodic point x i i 1 n , i n . Let q i N + be a period of x i , and set q = lcm ( q 1 , , q n ) . Define φ = i = 1 n x i χ J i . Since φ is measurable and takes values in I, we have φ L 1 ( I , I ) .
For every i { 1 , , n } , write q = i q i for some i N + . Since q i is a period of x i , we have
f 1 , r + q i ( x i ) = f 1 , r ( x i ) for every r N 0 .
For any m, applying this identity successively to r = m + ( i 1 ) q i , m + ( i 2 ) q i , , m , we obtain
f 1 , m + q ( x i ) = f 1 , m + i q i ( x i ) = f 1 , m + ( i 1 ) q i ( x i ) = = f 1 , m ( x i ) .
Therefore, for every t I , since t J i for some i, it follows that
H 1 , m + q ( φ ) ( t ) = f 1 , m + q ( φ ( t ) ) = f 1 , m + q ( x i ) = f 1 , m ( x i ) = H 1 , m ( φ ) ( t ) .
Hence
H 1 , m + q ( φ ) = H 1 , m ( φ ) for every m N 0 ,
and thus φ is a periodic point of the functional envelope system.
Moreover, for every t J i , both h ( t ) and x i belong to the interval i 1 n , i n , and hence | x i h ( t ) | 1 n . It follows that
ρ ( φ , h ) = I | φ ( t ) h ( t ) | d μ ( t ) = i = 1 n J i | x i h ( t ) | d μ ( t ) i = 1 n 1 n μ ( J i ) = 1 n μ ( I ) = 1 n .
Therefore,
ρ ( φ , g ) ρ ( φ , h ) + ρ ( h , g ) < ε 2 + ε 2 = ε .
Thus every open ball in L 1 ( I , I ) contains a periodic point of ( L 1 ( I , I ) , H ) . Hence the set of periodic points of the functional envelope system is dense in L 1 ( I , I ) .
(2) ⟹ (1). Suppose that the set of periodic points of ( L 1 ( I , I ) , H ) is dense in L 1 ( I , I ) . Let x I and ε > 0 be arbitrary, and set r = min ε , 1 2 . Consider the constant function
g x ( t ) = x , t I .
By the density of the periodic points of the functional envelope system, there exists a periodic point ψ L 1 ( I , I ) such that
ρ ( ψ , g x ) = I | ψ ( t ) x | d μ ( t ) < r 2 .
Let q N + be a period of ψ . Thus,
H 1 , m + q ( ψ ) = H 1 , m ( ψ ) in L 1 ( I , I )
for every m N 0 .
Choose a measurable representative of ψ , still denoted by ψ . Define A = t I : | ψ ( t ) x | < r . By Chebyshev’s inequality,
μ ( I A ) = μ t I : | ψ ( t ) x | r 1 r I | ψ ( t ) x | d μ ( t ) = ρ ( ψ , g x ) r < r 1 2 .
Hence μ ( A ) > 0 .
For each m N 0 , the equality H 1 , m + q ( ψ ) = H 1 , m ( ψ ) in L 1 ( I , I ) implies that there exists a measurable subset E m I with μ ( E m ) = 1 such that
f 1 , m + q ( ψ ( t ) ) = f 1 , m ( ψ ( t ) ) for every t E m .
Let E = m = 0 E m . Since this is a countable intersection of sets of full measure, μ ( E ) = 1 . Because μ ( A ) > 0 , the intersection A E is nonempty. Choose t 0 A E , and put y = ψ ( t 0 ) . Since t 0 A ,
| y x | = | ψ ( t 0 ) x | < r ε .
Furthermore, since t 0 E , for every m N 0 we have
f 1 , m + q ( y ) = f 1 , m + q ( ψ ( t 0 ) ) = f 1 , m ( ψ ( t 0 ) ) = f 1 , m ( y ) .
Thus y is a periodic point of ( I , f ) .
Since x I and ε > 0 were arbitrary, the periodic points of ( I , f ) are dense in I. □
Theorem 3. 
Let ( I , f ) be a nonautonomous discrete dynamical system on I = [ 0 , 1 ] , and let L 1 ( I , I ) , H be its L 1 -functional envelope system. Then the following statements are equivalent:
1. 
( I , f ) is weakly mixing of all orders;
2. 
L 1 ( I , I ) , H is topologically transitive;
3. 
L 1 ( I , I ) , H is weakly mixing of all orders.
Proof of Theorem 3. 
The implication ( 3 ) ( 2 ) follows from weak mixing of order 2. Indeed, let U , V be nonempty open subsets of L 1 ( I , I ) . Applying weak mixing of order 2 to the two pairs ( U , V ) and ( L 1 ( I , I ) , L 1 ( I , I ) ) , we obtain N H ( U , V ) . Hence the functional envelope system is topologically transitive.
(2) ⟹ (1). This is analogous to the proof of [10].
(1) ⟹ (3). Suppose that ( I , f ) is weakly mixing of all orders. Let m N + , and let U 1 , , U m , V 1 , , V m be arbitrary nonempty open subsets of L 1 ( I , I ) .
For each l = 1 , , m , choose
g l U l and h l V l .
Since the above sets are open and their number is finite, there exists ε > 0 such that
B ρ ( g l , ε ) U l and B ρ ( h l , ε ) V l
for every l = 1 , , m .
Choose n 2 such that 1 n < ε . For r = 1 , , n 1 , set
C r = r 1 n , r n , r = 1 , , n 1 , and C n = n 1 n , 1 .
Choose measurable representatives of g l and h l , still denoted by g l and h l , and define
L r , l = { t I : g l ( t ) C r } and J s , l = { t I : h l ( t ) C s } .
where l = 1 , , m , r , s = 1 , , n . For each l, both
L 1 , l , , L n , l and J 1 , l , , J n , l
are measurable partitions of I.
Let O r = r 1 n , r n , r = 1 , , n . Apply weak mixing of order n 2 to the collection of pairs ( O r , O s ) : 1 r , s n . Then there exists a common integer k N + such that
f 1 , k ( O r ) O s for all 1 r , s n .
For each pair ( r , s ) , choose x r , s O r such that f 1 , k ( x r , s ) O s .
For l = 1 , , m and r , s = 1 , , n , define W r , s l = L r , l J s , l . For each fixed l, the sets { W r , s l : 1 r , s n } form a measurable partition of I. Define
φ l = r = 1 n s = 1 n x r , s χ W r , s l .
Then φ l is measurable and takes values in I, and hence φ l L 1 ( I , I ) .
If t W r , s l , then
g l ( t ) C r and x r , s O r C r .
Consequently, | x r , s g l ( t ) | 1 n . It follows that
ρ ( φ l , g l ) = I | φ l ( t ) g l ( t ) | d μ ( t ) = r = 1 n s = 1 n W r , s l | x r , s g l ( t ) | d μ ( t ) 1 n r = 1 n s = 1 n μ ( W r , s l ) = 1 n μ ( I ) = 1 n < ε .
Thus,
φ l B ρ ( g l , ε ) U l .
Moreover, if t W r , s l , then
h l ( t ) C s and f 1 , k ( x r , s ) O s C s .
Therefore,
| f 1 , k ( x r , s ) h l ( t ) | 1 n .
Hence
ρ H 1 , k ( φ l ) , h l = I | f 1 , k ( φ l ( t ) ) h l ( t ) | d μ ( t ) = r = 1 n s = 1 n W r , s l | f 1 , k ( x r , s ) h l ( t ) | d μ ( t ) 1 n r = 1 n s = 1 n μ ( W r , s l ) = 1 n < ε .
Thus,
H 1 , k ( φ l ) B ρ ( h l , ε ) V l .
It follows that
H 1 , k ( U l ) V l for every l = 1 , , m .
Equivalently,
k l = 1 m N H ( U l , V l ) .
Since m and the open sets U l , V l were arbitrary, the functional envelope system ( L 1 ( I , I ) , H 1 , ) is weakly mixing of all orders. □
Lemma 3. 
Let L 1 ( I , I ) , H be an L 1 -functional envelope system. If it is weakly mixing of order 2, then it has sensitive dependence on initial conditions.
Proof of Lemma 3. 
Define the constant functions
0 ( t ) = 0 and 1 ( t ) = 1 , t I .
Clearly,
ρ ( 0 , 1 ) = I | 0 ( t ) 1 ( t ) | d μ ( t ) = 1 , 0 , 1 L 1 ( I , I ) .
Set δ = 1 4 , and let V 0 = B ρ ( 0 , δ ) , V 1 = B ρ ( 1 , δ ) . If α V 0 and β V 1 , then the triangle inequality gives
ρ ( α , β ) ρ ( 0 , 1 ) ρ ( α , 0 ) ρ ( β , 1 ) > 1 δ δ = 2 δ .
We prove that δ is a sensitivity constant. Let φ L 1 ( I , I ) and let ε > 0 . Put U = B ρ ( φ , ε ) . Since the functional envelope system is weakly mixing of order 2, applied to the two pairs ( U , V 0 ) and ( U , V 1 ) , there exists a common integer n N + such that
H 1 , n ( U ) V 0 , H 1 , n ( U ) V 1 .
Consequently, there exist φ 0 , φ 1 U such that
H 1 , n ( φ 0 ) V 0 and H 1 , n ( φ 1 ) V 1 .
Hence
ρ H 1 , n ( φ 0 ) , H 1 , n ( φ 1 ) > 2 δ .
On the other hand, by the triangle inequality,
ρ H 1 , n ( φ 0 ) , H 1 , n ( φ 1 ) ρ H 1 , n ( φ 0 ) , H 1 , n ( φ ) + ρ H 1 , n ( φ ) , H 1 , n ( φ 1 ) .
Therefore, at least one of the following inequalities holds:
ρ H 1 , n ( φ 0 ) , H 1 , n ( φ ) > δ , ρ H 1 , n ( φ 1 ) , H 1 , n ( φ ) > δ .
In the first case, set ψ = φ 0 , and in the second case, set ψ = φ 1 . Since φ 0 , φ 1 B ρ ( φ , ε ) , we obtain ψ B ρ ( φ , ε ) and
ρ H 1 , n ( φ ) , H 1 , n ( ψ ) > δ .
Thus L 1 ( I , I ) , H has sensitive dependence on initial conditions. □
Proof of Theorem 1. 
By Lemma 2, the assumption that ( I , f ) is weakly mixing of order 3 implies that it is weakly mixing of every order m 2 . Hence ( I , f ) is weakly mixing of all orders.
By Theorem 3, the functional envelope system L 1 ( I , I ) , H is weakly mixing of all orders. In particular, it is topologically transitive and weakly mixing of order 2. Therefore, by Lemma 3, it has sensitive dependence on initial conditions. On the other hand, since Per ( f ) is dense in I, Theorem 2 implies that Per ( H ) is dense in L 1 ( I , I ) .
Thus the functional envelope system L 1 ( I , I ) , H is topologically transitive, has a dense set of periodic points, and has sensitive dependence on initial conditions. Therefore, it is Devaney chaotic. □

Conflicts of Interest

The author declares no conflicts of interest.

References

  1. Kolyada, S.; Snoha, L. Topological entropy of nonautonomous dynamical systems. Random Comput. Dynam. 1996, 4, 205–233. [Google Scholar]
  2. Devaney, R.L. An Introduction to Chaotic Dynamical Systems, 2nd ed.; Addison-Wesley Publishing Company: Redwood City, CA, USA, 1989. [Google Scholar]
  3. Banks, J.; Brooks, J.; Cairns, G.; Davis, G.; Stacey, P. On Devaney’s definition of chaos. Amer. Math. Mon. 1992, 99, 332–334. [Google Scholar] [CrossRef]
  4. Chen, Z.; Huang, T.; Huang, Y.; Liu, X. Chaotic behaviors of one dimensional wave equations with van der Pol nonlinear boundary conditions. J. Math. Phys. 2018, 59, 022704. [Google Scholar] [CrossRef]
  5. Chu, H.; Kang, D.; Ku, S. On stabilities of functional envelopes and its applications. J. Differ. Equ. Appl. 2010, 16, 1017–1029. [Google Scholar] [CrossRef]
  6. Das, T.; Shah, E.; Snoha, L. (Non-)expansivity in functional envelopes. J. Math. Anal. Appl. 2014, 410, 1043–1048. [Google Scholar] [CrossRef]
  7. Kolyada, S.; Semikina, J. On topological entropy: When positivity implies +. Proc. Amer. Math. Soc. 2015, 143, 1545–1558. [Google Scholar]
  8. Chen, Z.; Huang, Y. Functional envelopes relative to the point-open topology on a subset. Discret. Contin. Dyn. Syst. 2017, 37, 1227–1246. [Google Scholar] [CrossRef]
  9. Gong, H.; Zhang, S.; Zhou, X. Distributional chaos of functional envelopes on Cantor spaces. Qual. Theory Dyn. Syst. 2026, accepted, 1568. [Google Scholar]
  10. Chen, Z.; Huang, Y.; Sun, H.; Zhou, T. Chaotic behaviors of one-dimensional wave equations with van der Pol boundary conditions containing a source term. Internat. J. Bifur. Chaos Appl. Sci. Engrg. 2020, 30, 2050227. [Google Scholar] [CrossRef]
  11. S’anchez, I.; Sanchis, M.; Villanueva, H. Chaos in hyperspaces of nonautonomous discrete systems. Chaos Solitons Fractals 2017, 94, 68–74. [Google Scholar] [CrossRef]
  12. Balibrea, F.; Oprocha, P. Weak mixing and chaos in nonautonomous discrete systems. Appl. Math. Lett. 2012, 25, 1135–1141. [Google Scholar] [CrossRef]
  13. Jiao, L.; Wang, H.; Wang, L.; Wang, N.; Furui, S. A Note on Stronger Forms of Sensitivity for Non-Autonomous Dynamical Systems on Uniform Spaces. Entropy 2024, 26, 47. [Google Scholar] [CrossRef] [PubMed]
  14. Echenausia-Monroy, J. L.; Ontanon-Garcia, L. J.; Magallon-Garcia, D. A.; Huerta-Cuellar, G.; Gilardi-Velazquez, H. E.; Cuesta-Garcia, J. R.; Rivera-Rodriguez, R.; Alvarez, J. The Shape of Chaos: A Geometric Perspective on Characterizing Chaos. Mathematics 2025, 14, 15. [Google Scholar] [CrossRef]
  15. Fatima, S.; Shah, W.; Mirza, H.; Ramirez, C.; Lopez, J.; Gilardi-Velazquez, H.; Reategui, R.; Huerta-Cuellar, G. Topological Machine Learning Framework for Phase Portrait Classification of Nonlinear Dynamical Systems. Mathematics 2026, 14, 1939. [Google Scholar] [CrossRef]
  16. Levicka, A.; Samuilik, I. Dynamics and Bifurcation Analysis of a Generalized Three-Dimensional Chaotic Financial System. Mathematics 2026, 14, 1154. [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.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.