J. Fixed Point Theory Appl.

(2020) 22:38
https://doi.org/10.1007/s11784-020-0769-9
Published online March 21, 2020 Journal of Fixed Point Theory
c Springer Nature Switzerland AG 2020 and Applications

Approximating fixed points of enriched

contractions in Banach spaces
Vasile Berinde and Mădălina Păcurar

Abstract. We introduce a large class of contractive mappings, called en-

riched contractions, a class which includes, amongst many other contrac-
tive type mappings, the Picard–Banach contractions and some nonex-
pansive mappings. We show that any enriched contraction has a unique
fixed point and that this fixed point can be approximated by means
of an appropriate Krasnoselskij iterative scheme. Several important re-
sults in fixed point theory are shown to be corollaries or consequences
of the main results of this paper. We also study the fixed points of local
enriched contractions, asymptotic enriched contractions and Maia-type
enriched contractions. Examples to illustrate the generality of our new
concepts and the corresponding fixed point theorems are also given.
Mathematics Subject Classification. Primary 47H05; Secondary 47H10,
54J25.
Keywords. Banach space, enriched contraction, fixed point, Krasnosel-
skij iteration, strong convergence, local enriched contraction, Maia-type
enriched contraction.

1. Introduction
Let X be a nonempty set and T : X → X a self mapping. We denote the
set of fixed points of T by Fix (T ), i.e., Fix (T ) = {a ∈ X : T a = a} and
define the nth iterate of T as usually, that is, T 0 = I (identity map) and
T n = T n−1 ◦ T , for n ≥ 1.
The mapping T is said to be a Picard operator, see, for example, Rus
[21], if (i) Fix (T ) = {p} and (ii) T n x0 → p as n → ∞, for any x0 in X.
The most useful class of Picard operators, that play a crucial role in
nonlinear analysis, is the class of mappings known in the literature as Picard–
Banach contractions, first introduced by Banach in [3], in the case of what
we call now a Banach space, and then extended to complete metric spaces
by Caccioppoli [14].

This work was completed with the support of our TEX-pert.

38 Page 2 of 10 V. Berinde and M. Păcurar JFPTA

The Banach contraction mapping principle essentially states that, in

a complete metric space (X, d), any contraction T : X → X, that is, any
mapping for which there exists c ∈ [0, 1) such that
d(T x, T y) ≤ c · d(x, y), ∀x, y ∈ X (1.1)
is a Picard operator.
Starting from the Picard–Banach fixed point theorem, a very impressive
literature has developed in the last nine decades or so, see, for example, the
monographs [5,15,22] and references therein. Moreover, the Picard–Banach
fixed point theorem itself and some of its extensions have become very useful
and versatile tools in solving various nonlinear problems: differential equa-
tions, integral equations, integro-differential equations, optimization prob-
lems, variational inequalities, etc., see [5,22,26] and [27] and the extensive
literature cited there.
Having in view both the theoretical and applicative importance of the
Banach contraction mapping principle as well as the results obtained in
[7,8,11] and [12], the aim of this paper is to introduce a larger class of Picard
operators, called enriched Banach contractions, that includes Picard–Banach
contractions as a particular case and also includes some nonexpansive map-
pings.
We then study the existence and uniqueness of fixed points of en-
riched Banach contractions and prove a strong convergence theorem for the
Kransnoselskij iteration used to approximate the fixed points of enriched Ba-
nach contractions. We also consider a local variant of the Picard–Banach fixed
point theorem and further extend the main results to the case of asymptotic
enriched Banach contractions. Examples to illustrate the generality of our
new results are also presented.

2. Approximating fixed points of enriched contractions

Definition 2.1. Let (X, · ) be a linear normed space. A mapping T : X → X
is said to be an enriched contraction if there exist b ∈ [0, +∞) and θ ∈ [0, b+1)
such that
b(x − y) + T x − T y ≤ θ x − y , ∀x, y ∈ X. (2.1)
To indicate the constants involved in (2.1), we shall also call T a (b, θ)-
enriched contraction.
Example 1.

(1) Any contraction T with contraction constant c is a (0, c)-enriched con-

traction, i.e., T satisfies (2.1) with b = 0 and θ = c ∈ [0, 1).
(2) Let X = [0, 1] be endowed with the usual norm and let T : X → X be
defined by T x = 1 − x, for all x ∈ [0, 1]. Then T is nonexpansive (it is
an isometry), T is not a contraction but T is an enriched contraction.
Indeed, if T would be a contraction then, by (1.1), there would exist
c ∈ [0, 1) such that
|x − y| ≤ c · |x − y|, ∀x, y ∈ [0, 1],
Vol. 22 (2020) Approximating fixed points of enriched contractions Page 3 of 10 38

which, for any x = y, leads to the contradiction 1 ≤ c < 1.

On the other hand, the enriched contraction condition (2.1) is in this
case equivalent to
|(b − 1)(x − y)| ≤ θ · |x − y|, ∀x, y ∈ [0, 1],
with θ ∈ [0, b + 1). The previous inequality holds true for all x, y ∈ [0, 1] if
one chooses 0 < b < 1 and θ = 1 − b.
Hence, foranyb ∈ (0, 1), T is a (b, 1 − b)-enriched contraction. Note also
1
that Fix (T ) = .
2
Remark 2.2. We note that for T in Example 2 (2), the Picard iteration {xn }
associated with T , that is, the sequence xn+1 = 1 − xn , n ≥ 0, does not
1
converge for any initial guess x0 different of , the unique fixed point to T .
2
This suggests us that to approximate fixed points of enriched contrac-
tions we need more elaborate fixed point iterative schemes, like the Kras-
noselskij iterative method [13], for which we prove in the following a strong
convergence theorem in the class of enriched contractions. But first we need
to recall an important property.

Remark 2.3. Considering a self-mapping T on a convex subset C of a linear

space X, then for any λ ∈ (0, 1), the so-called averaged mapping (a term
coined in [2]) Tλ given by
Tλ x = (1 − λ)x + λT x, ∀x ∈ C (2.2)
has the property that Fix( Tλ ) = Fix (T ).

Theorem 2.4. Let (X, · ) be a Banach space and T : X → X a (b, θ)-

enriched contraction. Then
(i) Fix (T ) = {p};
(ii) there exists λ ∈ (0, 1] such that the iterative method {xn }∞
n=0 , given by

xn+1 = (1 − λ)xn + λT xn , n ≥ 0, (2.3)

converges to p, for any x0 ∈ X;
(iii) the following estimate holds
ci
xn+i−1 − p ≤ · xn − xn−1 , n = 0, 1, 2, . . . ; i = 1, 2, . . . , (2.4)
1−c
θ
where c = .
b+1
Proof. We split the proof into two different cases.
1
Case 1. b > 0. In this case, let us denote λ = . Obviously, 0 < λ < 1
b+1
and the enriched contractive condition (2.1) becomes
  
 1 
 − 1 (x − y) + T x − T y  ≤ θ x − y , ∀x, y ∈ X,
 λ 
38 Page 4 of 10 V. Berinde and M. Păcurar JFPTA

which can be written in an equivalent form as

Tλ x − Tλ y ≤ c · x − y , ∀x, y ∈ X, (2.5)
where we denoted c = λθ and Tλ is the averaged mapping defined in (2.2).
Since θ ∈ (0, b + 1), it follows that c ∈ (0, 1) and, therefore, by (2.5) Tλ
is a c-contraction.
In view of (2.2), the Krasnoselskij iterative process {xn }∞ n=0 defined by
(2.3) is exactly the Picard iteration associated with Tλ , that is,
xn+1 = Tλ xn , n ≥ 0. (2.6)
Take x = xn and y = xn−1 in (2.5) to get
xn+1 − xn ≤ c · xn − xn−1 , n ≥ 1. (2.7)
From (2.7), one obtains routinely the following two estimates:
1 − cm
xn+m − xn ≤ cn · · x1 − x0 , n ≥ 0, m ≥ 1 (2.8)
1−c
and
1 − cm
xn+m − xn ≤ c · · xn − xn−1 , n ≥ 1, m ≥ 1. (2.9)
1−c
Now, by (2.8), it follows that {xn }∞
n=0 is a Cauchy sequence and hence it is
convergent in the Banach space (X, · ). Let us denote
p = lim xn . (2.10)
n→∞

By letting n → ∞ in (2.6) and using the continuity of Tλ , we immediately

obtain
p = Tλ p,
that is, p ∈ F ix (Tλ ).
Next, we prove that p is the unique fixed point of Tλ . Assume that q = p
is another fixed point of Tλ . Then by (2.5)
0< p−q ≤c· p−q < p−q ,
a contradiction. Hence, Fix (Tλ ) = {p} and since, by (2.6), Fix (T ) = Fix( Tλ ),
claim (i) is proven.
Conclusion (ii) follows by (2.10).
To prove (iii), we let m → ∞ in (2.8) and (2.9) to get
cn
xn − p ≤ · x1 − x0 , n≥1 (2.11)
1−c
and
c
xn − p ≤ · xn − xn−1 , n ≥ 1, (2.12)
1−c
θ
respectively, where c = . Now one can merge (2.11) and (2.12) to get
b+1
the unifying error estimate (2.4).
Case 2. b = 0. In this case, λ = 1, c = θ and we proceed like in Case
1 but with T (= T1 ) instead of Tλ , when the Krasnoselskij iteration (2.3)
reduces, in fact, to the simple Picard iteration associated with T ,
Vol. 22 (2020) Approximating fixed points of enriched contractions Page 5 of 10 38

xn+1 = T xn , n ≥ 0.


Remark 2.5.

(1) In the particular case b = 0, by Theorem 2.4, we get the classical Banach
contraction fixed point theorem (see [3]) in the setting of a Banach space.
(2) Enriched Kannan contractions have been introduced and studied in [11].
Although the mapping T in Example 2 (2) is simultaneously an enriched
Kannan mapping and an enriched contraction, it is easily seen that the
class of enriched Kannan contractions is also independent of the class of
enriched Banach contractions, due to the fact that, see, for example [19],
the class of Kannan contractions is independent of the class of Banach
contractions.
We shall call a mapping T to be a strictly enriched Kannan mapping
if it is not a usual Kannan mapping. So, an open problem would be to find
strictly enriched Kannan mappings which are not enriched contractions.

3. Local and asymptotic versions of enriched contraction

mapping principle
The Picard–Banach contraction mapping principle has a useful local version,
see, for example [16], which involves an open ball B in a complete metric space
(X, d) and a nonself contraction map of B into X which has the essential
property that it does not displace the centre of the ball too far. The analogue
of this result in the case of enriched contractions is given by the following
Corollary.

Corollary 3.1. Let (X, · ) be a Banach space, r > 0, B = B(x0 , r) := {x ∈

X : x − x0 < r} and let T : B → X be a (b, θ)-enriched contraction. If
T x0 − x0 < (b + 1 − θ)r, then T has a fixed point.

Proof. We can choose ε < r such that

T x0 − x0 ≤ (b + 1 − θ)ε < (b + 1 − θ)r.
On the other hand, as T is a (b, θ)-enriched contraction, there exist b ∈
[0, +∞) and θ ∈ [0, b + 1) such that
b(x − y) + T x − T y ≤ θ x − y , ∀x, y ∈ B.
1
For b > 0, we take λ = and in a similar manner to the proof of Theorem
b+1
θ
2.4, one obtains that Tλ is a c-contraction on B, with c = λθ = .
b+1
Note also that from
 the inequality
 T x0 −x0 ≤ (b+1−θ)ε, one obtains
θ
that Tλ x0 − x0 ≤ 1 − ε = (1 − c)ε.
b+1
38 Page 6 of 10 V. Berinde and M. Păcurar JFPTA

We prove now that the closed ball B := {x ∈ X : x − x0 ≤ ε} is

invariant under Tλ . Indeed, for any x ∈ B, we have
Tλ x − x0 ≤ Tλ x − Tλ x0 + Tλ x0 − x0
≤ c x − x0 + (1 − c)ε ≤ cε + (1 − c)ε = ε,
which proves that Tλ (x) ∈ B, for any x ∈ B.
Since B is complete, the conclusion follows by Theorem 2.4.
If b = 0, we proceed in a similar way but we shall use the contraction
mapping principle to get the final conclusion. 

Remark 3.2. In the particular case b = 0, by Corollary 3.1, we obtain the local
variant of the contraction mapping principle, see, for example, Corollary 1.2
in [16].

The following example shows that there exist mappings T which are not
contractions but a certain iterate of them is a contraction.

Example 2. (Example 1.3.1, [20]) Let X = R and T : X → X be given by

1
T x = 0, if x ∈ (−∞, 2] and T x = − , if x ∈ (2, +∞). Then T is not a
3
contraction (being discontinuous) but T 2 is a contraction.

In such a case, we cannot apply the classical Picard–Banach contraction

mapping principle and thus the following fixed point theorem is useful, see,
for example, Theorem 1.3.2 in [20].

Theorem 3.3. Let (X, d) be a complete metric space and let T : X → X be a

mapping. If there exists a positive integer N such that T N is a contraction,
then Fix (T ) = {x∗ }.

Our first aim in this section is to obtain a similar result for the more
general case of enriched contractions in the setting of a Banach space.

Theorem 3.4. Let (X, · ) be a Banach space and let U : X → X be a

mapping with the property that there exists a positive integer N such that
U N is a (b, θ)-enriched contraction. Then
(i) Fix (U ) = {p};
(ii) there exists λ ∈ (0, 1] such that the iterative method {xn }∞
n=0 , given by

xn+1 = (1 − λ)xn + λU N xn , n ≥ 0,
converges to p, for any x0 ∈ X.

Proof. We apply Theorem 2.4 (i) for the mapping T = U N and obtain that
F ix (U N ) = {p}. We also have
U N (U (p)) = U N +1 (p) = U (U N (p)) = U (p),
which shows that U (p) is a fixed point of U N . But U N has a unique fixed
point, p; hence, U (p) = p and so p ∈ F ix (U ).
The remaining part of the proof follows by Theorem 2.4. 
Vol. 22 (2020) Approximating fixed points of enriched contractions Page 7 of 10 38

One of the most interesting generalizations of the contraction mapping

principle is the so-called Maia fixed point theorem, see [17], which was ob-
tained by splitting the assumptions in the contraction mapping principle
among two metrics defined on the same set. Its statement reads as follows.
Theorem 3.5. (Theorem 1.3.10, [20]) Let X be a nonempty set, d and ρ two
metrics on X and T : X → X a mapping. Suppose that
(i) d(x, y) ≤ ρ(x, y), for each x, y ∈ X;
(ii) (X, d) is a complete metric space;
(iii) T : X → X is continuous with respect to the metric d;
(iv) T is a contraction mapping with respect to the metric ρ.
Then T is a Picard operator.
The next theorem is an extension of Theorem 3.5 to the class of enriched
contractions but is established in the particular case of a Banach space.
Theorem 3.6. Let X be a linear space, · d and · ρ two norms on X and
T : X → X a mapping. Suppose that
(i) x − y d ≤ x − y ρ , for each x, y ∈ X;
(ii) (X, · d ) is a Banach space;
(iii) T : X → X is continuous with respect to the norm · d ;
(iv) T is a (b, θ)-enriched contraction mapping with respect to the norm · ρ .
Then T is a Picard operator.
Proof. Let x0 ∈ X. By (iv), we deduce similar to the proof of Theorem 2.4
1
that {Tλn x0 } is a Cauchy sequence in (X, · ρ ), where as usually λ = .
b+1
By (i), {Tλ x0 } is a Cauchy sequence in (X, · d ) and by (ii) it converges.
n

Let
x∗ = lim Tλn x0 .
n→∞
By (iii) we obtain that x∗ ∈ F ix (Tλ ) and by (iv) that F ix (Tλ ) = {x∗ }. Since
F ix (Tλ ) = F ix (T ), the conclusion follows. 

4. Conclusions
1. We introduced the class of enriched contractions, which includes the
Picard–Banach contractions as a particular case and also certain non-
expansive mappings.
2. We have shown that any enriched contraction has a unique fixed point
that can be approximated by means of some Kransnoselskij iterations. In
particular, by our fixed point results established in this paper we obtain
the classical Banach contraction principle in the setting of a Banach
space.
3. We presented examples to show that the class of enriched contractions
strictly includes the Picard–Banach contractions in the sense that there
exist mappings which are not contractions and belong to the class of
enriched contractions.
38 Page 8 of 10 V. Berinde and M. Păcurar JFPTA

4. It is worth mentioning that enriched contractions preserve a fundamen-

tal property of Picard–Banach contractions, namely any enriched con-
traction has a unique fixed point and is continuous (one can easily show
this considering the definition).
5. It is well known that the class of Picard–Banach contractions is indepen-
dent of the class of Kannan contractions (which are in general discontin-
uous mappings) but, in view of Example 2, there exist enriched Banach
contractions which are simultaneously enriched Kannan contractions.
6. We also obtained a local version (Corollary 3.1) as well as an asymptotic
version (Theorem 3.3) of the enriched contraction mapping principle
(Theorem 2.4).
7. Last but not least, we obtained a fixed point theorem of Maia type
which extends the Maia fixed point theorem (Theorem 3.5) from the
class of Picard–Banach contractions to the class of enriched contractions
(Theorem 3.6).
8. Picard–Banach contractions have been previously extended in [4] to the
so called almost contractions, a class of mappings which preserves most
of the features of Picard–Banach contractions but may have more than
one fixed point, see also [1,6,10] and references therein. Our aim is to
unify in some future works the class of almost contractions and enriched
contractions and of some other classes of contractive mappings studied
in [9,18,23–25].

Publisher’s Note Springer Nature remains neutral with regard to jurisdic-

tional claims in published maps and institutional affiliations.

References
[1] Alghamdi, M., Berinde, V., Shahzad, N.: Fixed point of multivalued nonself
almost contractions. J. Appl. Math. 2013, Art. ID 621614 (2013)
[2] Baillon, J.B., Bruck, R.E., Reich, S.: On the asymptotic behavior of nonex-
pansive mappings and semigroups in Banach spaces. Houst. J. Math. 4(1), 1–9
(1978)
[3] Banach, S.: Sur les opérations dans les ensembles abstraits et leurs applications
aux équations intégrales. Fund. Math. 3, 133–181 (1922)
[4] Berinde, V.: Approximating fixed points of weak contractions using the Picard
iteration. Nonlinear Anal. Forum 9(1), 43–53 (2004)
[5] Berinde, V.: Iterative Approximation of Fixed Points. Lecture Notes in Math-
ematics, vol. 1912, 2nd edn. Springer, Berlin (2007)
[6] Berinde, V.: Approximating fixed points of implicit almost contractions. Hacet.
J. Math. Stat. 41(1), 93–102 (2012)
[7] Berinde, V.: Approximating fixed points of enriched nonexpansive mappings by
Krasnoselskij iteration in Hilbert spaces. Carpathian J. Math. 35(3), 293–304
(2019)
[8] Berinde, V.: Weak and strong convergence theorems for the Krasnoselskij it-
erative algorithm in the class of enriched strictly pseudocontractive operators.
Vol. 22 (2020) Approximating fixed points of enriched contractions Page 9 of 10 38

An. Univ. Vest Timiş. Ser. Mat.-Inform. 56(2), 13–27 (2018). https://doi.org/
10.2478/awutm-2018-0013
[9] Berinde, V., Petric, M.: Fixed point theorems for cyclic non-self single-valued
almost contractions. Carpathian J. Math. 31(3), 289–296 (2015)
[10] Berinde, V., Păcurar, M.: Iterative approximation of fixed points of single-
valued almost contractions. In: Fixed Point Theory and Graph Theory. Else-
vier, Amsterdam, pp. 29–97 (2016)
[11] Berinde, V., Păcurar, M.: Fixed point theorems for Kannan type mappings with
applications to split feasibility and variational inequality problems (submitted)
[12] Berinde, V., Păcurar, M.: Fixed point theorems for Chatterjea type mappings
in Banach spaces (submitted)
[13] Borwein, J., Reich, S., Shafrir, I.: Krasnoselski–Mann iterations in normed
spaces. Can. Math. Bull. 35(1), 21–28 (1992)
[14] Caccioppoli, R.: Un teorema generale sull’esistenza di elementi uniti in una
transformazione funzionale. Rend. Accad. Lincei 11, 794–799 (1930)
[15] Goebel, K., Reich, S.: Uniform convexity, hyperbolic geometry, and nonexpan-
sive mappings. In: Monographs and Textbooks in Pure and Applied Mathe-
matics, vol. 83. New York: Marcel Dekker, Inc. (1984)
[16] Granas, A., Dugundji, J.: Fixed Point Theory. Springer, New York (2003)
[17] Maia, M.: Un’osservazione sulle contrazioni metriche. Rend. Semin. Mat. Univ.
Padova 40, 139–143 (1968)
[18] Păcurar, M.: Iterative Methods for Fixed Point Approximation. Editura Riso-
print, Cluj-Napoca (2009)
[19] Rhoades, B.E.: A comparison of various definitions of contractive mappings.
Trans. Am. Math. Soc. 226, 257–290 (1977)
[20] Rus, I.A.: Generalized Contractions and Applications. Cluj-University Press,
Cluj-Napoca (2001)
[21] Rus, I.A.: Weakly Picard operators and applications. Semin. Fixed Point The-
ory Cluj-Napoca 2, 41–57 (2001)
[22] Rus, I.A., Petruşel, A., Petruşel, G.: Fixed Point Theory. Cluj University Press,
Cluj-Napoca (2008)
[23] Shioji, N., Suzuki, T., Takahashi, W.: Contractive mappings, Kannan mappings
and metric completeness. Proc. Am. Math. Soc. 126(10), 3117–3124 (1998)
[24] Subrahmanyam, P.V.: Remarks on some fixed-point theorems related to Ba-
nach’s contraction principle. J. Math. Phys. Sci. 8, 445–457 (1974). errata, ibid.
9 (1975), 195
[25] Suzuki, T.: Contractive mappings are Kannan mappings, and Kannan map-
pings are contractive mappings in some sense. Comment. Math. (Prace Mat.)
45(1), 45–58 (2005)
[26] Zeidler, E.: Nonlinear Functional Analysis and its Applications. III. Variational
Methods and Optimization. Springer, New York (1985)
[27] Zeidler, E.: Nonlinear Functional Analysis and its Applications. I. Fixed-Point
Theorems. Springer, New York (1986)
38 Page 10 of 10 V. Berinde and M. Păcurar JFPTA

Vasile Berinde
Department of Mathematics and Computer Science
North University Centre at Baia Mare Technical University of Cluj-Napoca
Victoriei 76
430122 Baia Mare
Romania
e-mail: vberinde@cunbm.utcluj.ro

Mădălina Păcurar
Department of Economics and Bussiness Administration in German Language Fac-
ulty of Economics and Bussiness Administration
Babeş-Bolyai University of Cluj-Napoca
T. Mihali 58-60
400591 Cluj-Napoca
Romania
e-mail: madalina.pacurar@econ.ubbcluj.ro