Available online at www.sciencedirect.

com

ScienceDirect

Indagationes Mathematicae xxx (xxxx) xxx

www.elsevier.com/locate/indag

Review

γ -Quantum product of white noise operators and

applications
Samah Horrigue ∗
University of Jeddah, College of Science, Department of Mathematics, Jeddah, Saudi Arabia
Department of Mathematics, Faculty of Sciences of Tunis, University of Tunis Almanar, Tunisia
Received 9 July 2018; received in revised form 4 February 2020; accepted 11 February 2020

Communicated by J.M.A.M. van Neerven

Abstract
The γ -quantum product of white noise operators which is the generalization of the γ -product, is
introduced on the basis of the analytic characterization theorem for operator symbols established within
the framework of white noise distribution theory. Existence and uniqueness of solutions are proved for
a certain class of ordinary differential equations for Fock space operators.
⃝c 2020 Royal Dutch Mathematical Society (KWG). Published by Elsevier B.V. All rights reserved.

Keywords: Fock space; White noise operator; Operator symbol; γ -product; Stochastic differential equation

Contents

1. Introduction....................................................................................................................... 2
2. Preliminary ....................................................................................................................... 3
2.1. White noise space ................................................................................................... 3
2.2. White noise operators .............................................................................................. 4
2.3. Wick product on white noises operators spaces ......................................................... 5
3. γ -quantum product of white noise operators ...................................................................... 5
3.1. γ -product on white noise space................................................................................ 5
3.2. γ -quantum product .................................................................................................. 6
3.3. γ -exponential function ............................................................................................. 8
4. Quantum–classical correspondence...................................................................................... 10
5. Applications ...................................................................................................................... 12
∗ Correspondence to: Department of Mathematics, Faculty of Sciences of Tunis, University of Tunis
Almanar, Tunisia.
E-mail address: samah.horrigue@fst.rnu.tn.

https://doi.org/10.1016/j.indag.2020.02.003
0019-3577/⃝ c 2020 Royal Dutch Mathematical Society (KWG). Published by Elsevier B.V. All rights reserved.

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
2 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx

5.1. Example: Convolution products of white noise space ................................................. 12

5.2. Applications: Ordinary quantum differential equation ................................................. 12
Acknowledgments.............................................................................................................. 14
References ........................................................................................................................ 14

1. Introduction

This paper aims at fusing the properties of infinite dimensional analysis and quantum theory
in order to generalize the wick product of white noise operators to γ -quantum product.
The white noise theory, which originally aims at extension of Itô theory keeping contact
with Lévy’s stochastic variational calculus [7], has developed considerably into an infinite
dimensional analysis with a many range of applications [8,16] and references therein. Among
others, the creation and annihilation operators (quantum white noise process) were formulated
as continuous operators on white noise functions sublimated into the white noise operator the-
ory [18]. They brought a new approach to quantum stochastic analysis [2,5,10,19]. Moreover,
together with the stochastic limit of quantum theory due to Accardi, Lu and Volovich [1],
which reveals a mechanism of emerging a quantum stochastic differential equation (in a broad
sense) from a standard Hamiltonian model, white noise approach is expected to be a clue to
go beyond the traditional Itô theory.
The main purpose of this paper is to give a first step toward a new theory of quantum
stochastic differential equations on the basis of the white noise distribution theory. We introduce
Q Q
the γ -quantum product of white noise operators by means of the Gα,β;γ ,δ and its adjoint Fα,β;γ ,δ .
Q
We recall that Gα,β;γ ,δ is the quantum generalized Fourier–Gauss transform which defined by
Q
Gα,β;γ ,δ (Ξ ) = Gα,γ Ξ Fβ,δ ,

where Gα,β -transform is characterized by

Gα,δ (ϕξ ) = exp(α⟨ξ, ξ ⟩)ϕδξ , ∀ξ ∈ E = S(R),
where ϕξ is the exponential function and S(R) is the Schwartz space and Fβ,δ is its adjoint.
Then, we discuss existence and uniqueness of a solution of a certain class of stochastic
differential equations which possess fairly singular coefficients.
This paper is organized as follows: In Section 2 we recall of well-known results in
white noise analysis. In Section 3, firstly, we recall from [9] the γ -product and give the
essentially properties. Then, we introduce a new concept of γ -quantum product which is the
noncommutative extensions of the γ -product on the white noise operators space. And we
will provide some properties and characterizations of γ -quantum product denoted ⋄γ . Then in
Section 5.2, we study the relations between the γ -quantum product and the classical γ -product.
Finally, in Section 5, we cite examples. We define the γ -exponential function. As application,
we resolve the following differential equation:
∂Ξ (t)
{
∂t
= At ⋄γ Ξ (t) + Bt , t ∈ [0, T ] ,
Ξ (0) = Ξ0 ∈ (E) ,
where At and Bt for t ∈ [0, T ] are a continuous L((E), (E)∗ )-process.

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx 3

2. Preliminary
2.1. White noise space

Let H be a complex Hilbert space with norm |.|0 . Let A be a selfadjoint operator in H
such that ∥A−1 ∥ O P < 1 and ∥A−1 ∥ H S < ∞, for more details see [16,17]. By the standard
construction from H and A, we have a Gelfand triple:
E ⊂ H ⊂ E ∗, (2.1)
∗
where E is the strong dual space of E. For each p ∈ R, we define the norm |.| p by
|ξ | = |A p ξ |0 , for any ξ ∈ E. Then E becomes a countable Hilbert nuclear space. The canonical
bilinear form on E ∗ × E is denoted by the same symbol ⟨., .⟩. The prototype of (2.1) is
E = S(C) ⊂ H = L 2 (C, dt) ⊂ E ∗ = S ′ (C),
where S(C) is the space of rapidly decreasing C ∞ -functions and S ′ (C) the space of tempered
distributions.
It follows from the Bochner–Minlos Theorem that there exists a probability measure µ on
E R∗ = S ′ (R), the dual space of the space S(R) of real-valued rapidly decreasing C ∞ -functions,
such that its characteristic function is given by
∫
− 21 |ξ |20
e = ei⟨x,ξ ⟩ µ(d x), for all ξ ∈ E R = S(R).
∗
ER

This µ is called the standard Gaussian measure. We denoted by L 2 (E R∗ , µ) the Hilbert space
of C-valued L 2 -functions on E R∗ where (E R∗ , µ) is called the Gaussian space. It is well known
that L 2 (E ∗ , µ)∑is unitary isomorphic to the Boson Fock space Γ (H ) = {φ = ( f n ); f n ∈
ˆ
H ⊗n , ∥φ∥2 = ∞ n=0 n!| f n |0 < ∞}, for more details see [8,16,18].
2

The second quantization of the operator A denoted Γ (A) is defined by

( )∞
Γ (A)φ = A⊗n f n , φ = ( f n )∞
n=0 ∈ Γ (H ).
n=0

Then Γ (A) is a selfadjoint operator in Γ (H ) with ∥Γ (A)−1 ∥ O P < 1 and ∥Γ (A)−1 ∥ H S < ∞.
We define the following space
(E) = proj limΓ (E p ), (E)∗ = ind limΓ (E − p ),
p→∞ p→∞

where {Γ (E p ); p ∈ Z} are an increasing chain of Fock spaces given by

∞
∑
ˆ
p , ∥φ∥ p = n!| f n |2p < ∞},
2
Γ (E p ) = {φ = ( f n ); f n ∈ E ⊗n
n=0
ˆ
where | f n | p is the norm of the nth symmetric tensor power E ⊗n
p . So, we construct the
Hida–Kubo–Takenaka space:
(E) ⊂ Γ (H ) ⊂ (E)∗
in the standard manner, for more details see e.g., [15,16,18]. The canonical C-bilinear form on
(E) × (E)∗ takes the form:
∞
∑
⟨⟨Φ, φ⟩⟩ = n!⟨Fn , f n ⟩, Φ = (Fn ) ∈ (E)∗ , φ = ( f n ) ∈ (E).
n=0

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
4 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx

2.2. White noise operators

We denoted by L((E), (E)∗ ) the space of the continuous operator from (E) into (E)∗ and
is equipped with the bounded convergence topology which is called the white noise operators
space. The annihilation operator at at a point t ∈ [0, T ] is defined by
at : (0, . . . , 0, ξ ⊗n , 0, . . .) ↦→ (0, . . . , 0, nξ (t)ξ ⊗(n−1) , 0, . . .), ξ ∈ E.
It is shown that at ∈ L((E), (E)). The creation operator at∗ at t is the adjoint operator of at .
The pair {at , at∗ ; t ∈ T } is called the quantum white noise on T . For more detailed account
of quantum white noise calculus see e.g., [13,18].
It is then proved in [18] that for any distribution κl,m ∈ (E ⊗(l+m) )∗ , an integral kernel
operator expressed in a formal integral:
∫
Ξl,m (κl,m ) = κl,m (s1 , . . . , sl , t1 , . . . , tm )
T l+m
as∗1 · · · as∗l at1 · · · atm ds1 · · · dsl dt1 · · · dtm
becomes a white noise operator. Moreover, Ξl,m (κl,m ) belongs to L((E), (E)) if and only if
κl,m ∈ (E ⊗l ) ⊗ (E ⊗m )∗ . Note that every white noise operator Ξ ∈ L((E), (E)∗ ) admits a Fock
expansion:
∞
∑
Ξ = Ξl,m (κl,m ).
l,m=0

For ζ ∈ E C∗ , we define white noise operators:

∫ ∫
a(ζ ) = Ξ0,1 (ζ ) = ζ (t)at dt, a (ζ ) = Ξ1,0 (ζ ) =
∗
ζ (t)at∗ dt, (2.2)
T T
which are called respectively the annihilation and creation operators associated with ζ .
Moreover, for ζ ∈ E, both a(ζ ) and a ∗ (ζ ) belong to L((E), (E)) ∩ L((E)∗ , (E)∗ ).
Let τ be the trace on H , i.e., ⟨τ, ξ ⊗ η⟩ = ⟨ξ, η⟩ for ξ, η ∈ E. Then τ ∈ (E ⊗2 )2 . Then, the
integral kernel operator with trace τ as kernel distribution,
∫
∆G = τ (s, t)at as dtds, (2.3)
T×
is called the Gross Laplacian. It is known that ∆G is continuous linear operator from (E) into
itself.
For Ξ ∈ L((E), (E)∗ ) the symbol of a white noise operator Ξ is defined by
σ (Ξ )(ξ, η) = Ξ̂ (ξ, η) = ⟨⟨Ξ φξ , φη ⟩⟩, ξ, η ∈ E, (2.4)
where φξ , for each ξ ∈ E, is the exponential vector associated with ξ ∈ E defined by
ξ ⊗2 ξ ⊗n
1, ξ,,... , . . . ∼ φξ ∈ (E).
( )
2! n!
Every white noise operator is uniquely determined by its symbol [4,17,18].
As examples, the symbols of Γ (A) and ∆G are given resp., by
σ (Γ (A))(ξ, η) = exp{⟨Aξ, η⟩} (2.5)
σ (∆G )(ξ, η) = ⟨ξ, ξ ⟩ exp{⟨ξ, η⟩}, (2.6)
for all ξ, η ∈ E.

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx 5

Then, the following theorem is well-known as analytic characterization theorem for symbols
of white noise operators.

Theorem 2.1 ([17]). A function Θ : E × E → C is the symbol of an operator Ξ ∈ L((E), (E)∗ )

if and only if
(i) Θ is a Gâteaux-entire function, i.e., for any ξ, ξ1 , η, η1 ∈ E, the function Θ(zξ +ξ1 , wη+η1 )
is entire holomorphic of (z, w) ∈ C × C.
(ii) there exist nonnegative constants C > 0 and p ≥ 0 such that
|Θ(ξ, η)| ≤ C exp{|ξ |2p + |η|2p }, ξ, η ∈ E. (2.7)

2.3. Wick product on white noises operators spaces

For Φ, Ψ ∈ (E)∗ , it is known that the Wick product Φ ⋄ Ψ ∈ (E)∗ is defined by

S(Φ ⋄ Ψ ) = SΦ · SΨ . We recall that the S-transform for any Φ function on (E) is defined by
SΦ(ξ ) = ⟨⟨Φ, φξ ⟩⟩, Φ ∈ (E)∗ , ξ ∈ E. (2.8)
Note that the Wick product is a continuous binary operation on (E), see e.g [16,18].
The Wick product is generalized to the white noise operators space in [20] as follows.

Definition 2.1. For each Ξ1 and Ξ2 in L((E), (E)∗ ), we define the Wick product, denoted
Ξ1 ⋄ Ξ2 , by:
σ (Ξ1 ⋄ Ξ2 )(ξ, η) = σ (Ξ1 )(ξ, η)σ (Ξ2 )(ξ, η)e−⟨ξ,η⟩ , ξ, η ∈ E. (2.9)
Then, we have the following properties:

Proposition 2.2 ([20]). For all Ξ , Ξ1 , Ξ2 ∈ L((E), (E)∗ ), it holds that:

1. Ξ ⋄ I = I ⋄ Ξ = Ξ,
2. Ξ ⋄ (Ξ1 ⋄ Ξ2 ) = (Ξ ⋄ Ξ1 ) ⋄ Ξ2 ,
3. (Ξ1 ⋄ Ξ2 )∗ = Ξ1∗ ⋄ Ξ2∗ ,
4. at∗ Ξ = at∗ ⋄ Ξ = Ξ ⋄ at∗ and at∗ as∗ = at∗ ⋄ as∗ ,
5. at Ξ = at ⋄ Ξ = Ξ ⋄ at and at as = at ⋄ as ,
6. as∗ at = as∗ ⋄ at and at∗ as = as ⋄ at∗ .

3. γ -quantum product of white noise operators

Throughout this paper, let γ be fixed but arbitrary complex number.
Firstly, we recall some definitions and properties from [9].

3.1. γ -product on white noise space

For α, β ∈ C, the Gα,β -transform characterized by

Gα,β (ϕξ ) = exp(α⟨ξ, ξ ⟩)ϕβξ , ∀ξ ∈ E, (3.1)
is well defined on (E) into itself since the {ϕξ , ξ ∈ E} spans a dense linear subspace of (E).
Fore more details, see [16].
∗
The adjoint operator Gα,β ∈ L((E)∗ , (E)∗ ) of Gα,β is denoted by Fα,β and is called the
Fα,β -transform.

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
6 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx

Definition 3.1 ([9]).

1. The γ -product, denoted by ⋄γ , is defined by
ϕ1 ⋄γ ϕ2 = G γ ,1 (G− γ ,1 (ϕ1 ) ⋄ G− γ ,1 (ϕ2 )), ϕ1 , ϕ2 ∈ (E). (3.2)
2 2 2

We can easily check that ⋄0 is the Wick product and ⋄1 is the Wiener product. It is easy
to see that:
ϕξ ⋄γ ϕη = eγ ⟨ξ,ξ ⟩ ϕξ +η , ∀ ξ, η ∈ E. (3.3)
2. For any Φ ∈ (E) and ϕ ∈ (E), we generalize the γ -product as follows:
∗

⟨⟨Φ ⋄γ ϕ, ψ⟩⟩ = ⟨⟨Φ, ϕ ⋄γ ψ⟩⟩, ∀ψ ∈ (E). (3.4)

Proposition 3.1 ([9]). For any Φ ∈ (E)∗ and ϕ ∈ (E), we have

Φ ⋄γ ϕ = F −γ ,1 (F γ ,1 (Φ) ⋄ G− γ ,1 (ϕ)). (3.5)
2 2 2

We can generalize the γ -product on the distribution space (E)∗ as follows:

Definition 3.2 ([9]). For any Φ, Ψ ∈ (E)∗ , the generalized γ -product on (E)∗ is defined by:
⟨⟨Φ ⋄γ Ψ , ϕ⟩⟩ = ⟨⟨Φ, Ψ ⋄γ ϕ⟩⟩, ∀ϕ ∈ (E). (3.6)

Theorem 3.2 ([9]). For all Φ, Ψ ∈ (E)∗ , the generalized γ -product of Φ and Ψ is given by
Φ ⋄γ Ψ = F −γ ,1 (F −γ ,1 Φ ⋄ F γ ,1 Ψ ). (3.7)
2 2 2

Moreover, we have
γ
S(Φ ⋄γ Ψ )(ξ ) = e− 2 ⟨ξ,ξ ⟩ S(Φ)(ξ )S(Ψ )(ξ ), ∀ξ ∈ E. (3.8)

Remark 3.1. It is easy to see from (3.8) that

Φ ⋄γ Ψ = F −γ ,1 (Φ ⋄ Ψ ). (3.9)
2

3.2. γ -quantum product

We recall the quantum generalized Fourier–Gauss transform from [11].

Definition 3.3. Let α, β, γ , δ ∈ C. Then for every Ξ ∈ L((E)∗ , (E)), we define the quantum
generalized Fourier–Gauss transform is defined by
Q
Gα,β;γ ,δ (Ξ ) = Gα,γ Ξ Fβ,δ . (3.10)
And the quantum generalized Fourier–Mehler transform is defined by
Q
Fα,β;γ ,δ (Ξ ) = Fα,γ Ξ Gβ,δ , Ξ ∈ L((E), (E)∗ ). (3.11)

Theorem 3.3 ([11]). The quantum generalized Fourier–Gauss transform, resp. the quantum
generalized Fourier–Mehler transform is well defined in L((E)∗ , (E)), resp. L((E), (E)∗ ), into
itself.

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx 7

Q Q
As Gα,β;γ ,δ and Fα,β;γ ,δ are well defined, we can define the γ -Quantum product as follows.

Definition 3.4. We define the γ -Quantum product of two operators Ξ1 , Ξ2 ∈ L((E), (E)∗ ),
denoted ⋄γ , by
Ξ1 ⋄γ Ξ2 = F γQ, γ ;1,1 (F−Qγ ,− γ ;1,1 (Ξ1 ) ⋄ F−Qγ ,− γ ;1,1 (Ξ2 )). (3.12)
2 2 2 2 2 2

Proposition 3.4. For all Ξ1 , Ξ2 ∈ L((E), (E)∗ ), we have:

Ξ1 ⋄γ Ξ2 = F γ ,1 F− γ ,1 (Ξ1 )G− γ ,1 ⋄ F− γ ,1 (Ξ2 )G− γ ,1 G γ ,1 .

( )
(3.13)
2 2 2 2 2 2

Moreover, we have Ξ1 ⋄γ Ξ2 ∈ L((E), (E)∗ ) and for all ξ, η ∈ E, we have

γ(
{ ) }
σ (Ξ1 ⋄γ Ξ2 )(ξ, η) = exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ − ⟨ξ, η⟩ σ (Ξ1 )(ξ, η)σ (Ξ2 )(ξ, η). (3.14)
2

Proof. Eq. (3.13) is deduced directly by using Eqs. (3.11) and (3.12).
For all Ξ1 , Ξ2 ∈ L((E), (E)∗ ) and for all ξ, η ∈ E, we have:

σ (Ξ1 ⋄γ Ξ2 )(ξ, η) = ⟨⟨Ξ1 ⋄γ Ξ2 (φξ ), φη ⟩⟩

= ⟨⟨F γ ,1 F− γ ,1 (Ξ1 )G− γ ,1 ⋄ F− γ ,1 (Ξ2 )G− γ ,1 G γ ,1 (φξ ), φη ⟩⟩
( )
2 2 2 2 2 2
γ
⟨ξ,ξ ⟩
( )
= e2 ⟨⟨ F− γ ,1 (Ξ1 )G− γ ,1 ⋄ F− γ ,1 (Ξ2 )G− γ ,1 (φξ ), G γ ,1 (φη )⟩⟩
2 2 2 2 2

γ
( )
= e 2 ⟨ξ,ξ ⟩+⟨η,η⟩−⟨ξ,η⟩ ⟨⟨F− γ ,1 (Ξ1 )G− γ ,1 (φξ ), φη ⟩⟩⟨⟨F− γ ,1 (Ξ2 )G− γ ,1 (φξ ), φη ⟩⟩
2 2 2 2

γ
{ ( ) }
= exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ − ⟨ξ, η⟩ σ (Ξ1 )(ξ, η)σ (Ξ2 )(ξ, η).
2
It follows by using Theorem 2.1 that for i = 1, 2 there exist K i > 0 and pi ∈ N such that

|σ (Ξ1 ⋄γ Ξ2 )(ξ, η)|

γ(
{ ) } {( ) ( )}
≤ K 1 K 2 | exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ − ⟨ξ, η⟩ | exp |ξ | p1 + |η| p1 + |ξ | p2 + |η| p2 .
2 2 2 2
2
On the other hand,
2 2 2 2 2
|e−⟨ξ,η⟩ | ≤ e2|ξ |0 |η|0 ≤ e|ξ |0 +|η|0 ≤ e|ξ | p +|η| p ,

where p = max{ p1 , p2 }. Then, we obtain

{ )}
|γ | + 5 ( 2
|σ (Ξ1 ⋄γ Ξ2 )(ξ, η)| ≤ K 1 K 2 exp |ξ | p + |η|2p .
2

Since |ξ | p ≤ ρ q |ξ | p+q , choose q ≥ 0 such that ρ 2q |γ |+5

2
≤ 1 so that
{( ) }
|σ (Ξ1 ⋄γ Ξ2 )(ξ, η)| ≤ K exp |ξ |2p+q + |η|2p+q , (3.15)

where K = K 1 K 2 , which complete the proof. □

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
8 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx

Corollary 3.5. For all Ξ , Ξ1 , Ξ2 ∈ L((E), (E)∗ ), it holds that:

1. Ξ ⋄γ Ω = Ω ⋄γ Ξ = Ξ , where Ω is an operator defined by
γ γ
Ω = ex p{ ∆∗G }Γ (I )ex p{ ∆G },
2 2
where ∆∗G is the adjoint of ∆G .
2. Ξ ⋄γ (Ξ1 ⋄γ Ξ2 ) = (Ξ ⋄γ Ξ1 ) ⋄γ Ξ2 .
3. (Ξ1 ⋄γ Ξ2 )∗ = Ξ1∗ ⋄γ Ξ2∗ .

Proof. 1. Let Ω ∈ L((E), (E)∗ ) such that:

Ξ ⋄γ Ω = Ω ⋄γ Ξ = Ξ .
It follows from (3.14), for all Ξ ∈ L((E), (E)∗ ), that:
γ(
{ ) }
σ (Ξ )(ξ, η) = exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ − ⟨ξ, η⟩ σ (Ω )(ξ, η)σ (Ξ )(ξ, η)
2
γ
{ ( ) }
⇔ σ (Ω )(ξ, η) = exp ⟨ξ, ξ ⟩ + ⟨η, η⟩ + ⟨ξ, η⟩ ∀η, ξ ∈ E.
2
2. and 3. are immediately proved by using (3.14). □

3.3. γ -exponential function

For simplicity, we put for n ∈ N:

n times
  
Ξ ⋄γ n
= Ξ ⋄γ · · · ⋄γ Ξ and Ξ ⋄γ 0 = I.

Definition 3.5 ([6]). Let α = {α(n)} be a positive sequence satisfying the following conditions:
(A1) α(0) = 1 and infn≥0 α(n)σ n > 0 for some σ ≥ 1;
(A2) limn→∞ { α(n)
n!
} = 0;
(A3) α is equivalent to a positive sequence γ = {γ (n)} such that { γn!(n) } is log-concave.
(A4) there exists a constant C1α > 0 such that α(m)α(n) ≤ C1α n+m
α(m + n) for all integer m
and n.
We define the complex weighted Fock space Wα = Γα (E) by:
∞
∑
Wα = proj limΓα (E p ) = proj lim{φ = ( f n ) ∈ (E) such that n!α(n)| f n |2p < ∞}.
p→∞ p→∞
n=0
(3.16)
We denote by Wα∗ the strong dual topology Wα which is given by
Wα∗ ∼
= ind limΓα−1 (E − p ),
p→∞

where ∼
= stands for a topological isomorphism. Then, by taking the complexification, we obtain
a complex Gelfand triple:
Wα ≡ Γα (E) ⊂ Γ (H ) ⊂ Γα (E)∗ ≡ Wα∗ .

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx 9

Theorem 3.6 ([6]). The symbol Θ = Ξ̂ of a white noise operator Ξ ∈ L(Wα , Wα∗ ) possesses
the following properties:
(O1) Θ is Gâteau-entire,
(O2) there exist constant numbers C > 0 and p ≥ 0 such that
|Θ(ξ, η)|2 ≤ C G α (|ξ |2p )G α (|η|2p ), ∀ξ, η ∈ E.

Theorem 3.7. Let Ξ ∈ L((E), (E)∗ ). For any complex number γ , the infinite series
∞
∑ 1 ⋄γ n
Ξ (3.17)
n=0
n!
converges in L(Wα , Wα∗ ), where α weight sequences satisfying (A1)–(A4) defined by his
generating function G α given by
∞
∑ α(n) |γ |+3 |γ | + 1
G α (t) = t n = exp(e 2 t + t − 1). (3.18)
n=0
n! 2

Proof. By induction, it is easy to see, for all integers n, that

( γ
{ }
σ (Ξ ⋄γ n )(ξ, η) = exp (n − 1) − (⟨ξ, ξ ⟩ + ⟨η, η⟩) − ⟨ξ, η⟩ σ (Ξ )n (ξ, η),
)
2
for all ξ, η ∈ E.
Let N be a positive integer and S N be the N th partial sum of (3.17). Using the definition
of symbol of operator and the linearity property, we obtain, for all ξ, η ∈ E, that:
N (ξ, η)
Sˆ
{( γ N
1 n − γ (⟨ξ,ξ ⟩+⟨η,η⟩)−⟨ξ,η⟩
)} ∑ {( )}
= e 2 (⟨ξ,ξ ⟩+⟨η,η⟩)+⟨ξ,η⟩ e 2 σ (Ξ )n (ξ, η).
n=0
n!
Then by a similar argument as in the proof of Proposition 3.4, it is easily verified that there
exist q ∈ N and K > 0 such that
N (ξ, η)| ≤ K G α (|ξ |q )G α (|η|q ), N = 1, 2, . . . .
2 2
| Sˆ
On the other hand, we have
lim SˆN (ξ, η)
N →∞
(γ
{ }
γ
(⟨ξ, ξ ⟩ + ⟨η, η⟩) + ⟨ξ, η⟩ + e− 2 (⟨ξ,ξ ⟩+⟨η,η⟩)−⟨ξ,η⟩ σ (Ξ )(ξ, η) .
)
= exp
2
Then, applying Theorem 2.1, we complete the proof. □

Remark 3.2. For γ = 0, the same results are proved in [6,14].

The convergent series introduced in Theorem 3.7 is called the γ -exponential function of Ξ
and is denoted by
∞
∑ 1 ⋄γ
expγ Ξ = Ξ . (3.19)
n=0
n!

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
10 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx

Moreover, the symbol of expγ Ξ is given by:

[γ γ
]
σ (expγ Ξ )(ξ, η) = exp (⟨ξ, ξ ⟩ + ⟨η, η⟩) + ⟨ξ, η⟩ + e− 2 (⟨ξ,ξ ⟩+⟨η,η⟩)−⟨ξ,η⟩ Ξ̂ (ξ, η) ,
2
(3.20)
for all ξ, η ∈ E.

Lemma 3.8. For all operators Ξ1 , Ξ2 ∈ L((E), (E)∗ ), we have

expγ (Ξ1 ) ⋄γ expγ (Ξ2 ) = expγ (Ξ1 + Ξ2 ). (3.21)
In particular, for any Ξ ∈ L((E), (E)∗ ), we have
expγ (Ξ ) ⋄γ expγ (−Ξ ) = I. (3.22)

Proof. For all Ξ1 , Ξ2 ∈ L((E), (E)∗ ) and ξ, η ∈ E, using (3.14) and (3.20), we have:
σ (expγ (Ξ1 ) ⋄γ expγ (Ξ2 ))(ξ, η)
γ
{[ ]}
− γ2 (⟨ξ,ξ ⟩+⟨η,η⟩)−⟨ξ,η⟩ ˆ
= exp (⟨ξ, ξ ⟩ + ⟨η, η⟩) + ⟨ξ, η⟩ + e (Ξ1 (ξ, η) + Ξ2 (ξ, η))
ˆ
2
= σ (expγ (Ξ1 ) + expγ (Ξ2 ))(ξ, η),
which is proved in Eq. (3.21).
To check (3.22), it is enough to replace Ξ1 by Ξ and Ξ2 by I in Eq. (3.21). □
Moreover, we can prove the following lemma without difficulty.

Lemma 3.9. Assume that Ξ ∈ L((E), (E)∗ ), then z ↦→ expγ (zΞ ) is entire holomorphic and
d
expγ (zΞ ) = Ξ ⋄γ expγ (zΞ ) (3.23)
dz
holds in L(Wα , Wα∗ ), where the generating function G α of the weighted sequence α is given
as in (3.18).

4. Quantum–classical correspondence
It is well known [16,18] that the pointwise multiplication yields a continuous bilinear map
from (E) × (E) into (E), i.e., for all f, g ∈ (E), we have f g ∈ (E). Let Φ ∈ (E)∗ and
MΦ ∈ L((E), (E)∗ ) be the multiplication operator by Φ by
⟨⟨MΦ f, g⟩⟩ = ⟨⟨Φ, f g⟩⟩, f, g ∈ (E). (4.1)
It follows that, for all ξ, η ∈ E, that:
σ (MΦ )(ξ, η) = e⟨ξ,η⟩ S(Φ)(ξ + η).
Let φ0 = (1, 0, . . .) ∈ (E) be the vacuum vector. Using the definition (4.1), we have the
following connection between the multiplication operator and the distribution Φ is given by
MΦ φ0 = Φ.
We recall from [11], the following results.

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx 11

Theorem 4.1. Let Φ ∈ (E ∗ ). Then, we have

( Q
Fα,β;γ ,δ (MΦ ) φ0 = Fα,γ (Φ).
)
(4.2)

Therefore, we obtain the following relation between the classical and the quantum
γ -product.

Theorem 4.2. For any Φ1 , Φ2 ∈ (E ∗ ), we have

MΦ1 ⋄γ MΦ2 = MΦ1 ⋄γ Φ2 . (4.3)

Then,
( )
MΦ1 ⋄γ MΦ2 φ0 = Φ1 ⋄γ Φ2 . (4.4)

Proof. To prove (4.3), we need only to show that the symbol of MΦ1 ⋄γ MΦ2 verifies the
following equation, for all ξ, η ∈ E:

σ (MΦ1 ⋄γ MΦ2 )(ξ, η) = σ (F−Qγ ,− γ ;1,1 (MΦ1 ⋄Φ2 ))(ξ, η).

2 2

Firstly, using (3.14), we have

σ (MΦ1 ⋄γ MΦ2 )(ξ, η)

γ(
{ ) }
= exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ − ⟨ξ, η⟩ σ (MΦ1 )(ξ, η)σ (MΦ2 )(ξ, η)
2
γ(
{ ) }
= exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ − ⟨ξ, η⟩ e⟨ξ,η⟩ S(Φ1 )(ξ + η)e⟨ξ,η⟩ S(Φ2 )(ξ + η)
2
γ(
{ )}
= exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ e⟨ξ,η⟩ S(Φ1 )(ξ + η)S(Φ2 )(ξ + η)
2
γ(
{ )}
= exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ S(Φ1 ⋄ Φ2 )(ξ + η). (4.5)
2
On the other hand, we have

σ (F−Qγ ,− γ ;1,1 (MΦ1 ⋄Φ2 ))(ξ, η) = ⟨⟨F− γ ,1 MΦ1 ⋄Φ2 G− γ ,1 φξ , φη ⟩⟩

2 2 2 2

γ
{ ( )}
= exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ σ (MΦ1 ⋄Φ2 )(ξ, η)
2
γ(
{ )}
= exp − ⟨ξ, ξ ⟩ + ⟨η, η⟩ S(Φ1 ⋄ Φ2 )(ξ + η). (4.6)
2
Eqs.(4.5) and (4.6) complete the proof of (4.3).
Using (4.3), (3.9) and (4.2), we obtain
( )
MΦ1 ⋄γ MΦ2 φ0 = F−Qγ ,− γ ;1,1 (MΦ1 ⋄Φ2 )φ0
2 2

= F− γ ,1 (Φ1 ⋄ Φ2 ) = Φ1 ⋄γ Φ2 . □
2

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
12 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx

5. Applications
5.1. Example: Convolution products of white noise space

We recall from [12] that for each fixed Υ ∈ L((E), (E)∗ ), we define a Υ -convolution
Ξ1 ∗Υ Ξ2 ∈ L((E), (E)∗ ) of two white noise operators Ξ1 , Ξ2 ∈ L((E), (E)∗ ) by
Ξ1 ∗Υ Ξ2 = Ξ1 ⋄ Ξ2 ⋄ Υ . (5.1)

Proposition 5.1. For all Ξ1 , Ξ2 ∈ L((E), (E)∗ ) and ξ, η ∈ E, we have

σ (Ξ1 ∗Υ Ξ2 )(ξ, η) = Ξ
ˆ1 (ξ, η)Ξ
ˆ2 (ξ, η)Υ̂ (ξ, η)e−2⟨ξ,η⟩ . (5.2)

Proof. The proof consisted in using the property (2.9). □

It provides the following result:

Proposition 5.2. For each fixed Υ ∈ L((E), (E)∗ ), the Υ -convolution products of white noise
space coincide with the γ -product if and only if
1 ∗ 1
Υ = e− 2 ∆G Γ (I )e− 2 ∆G .

Proof. Using (3.14) and (5.2) for all Ξ1 , Ξ2 ∈ L((E), (E)∗ ) and ξ, η ∈ E, we obtain
σ (Ξ1 ∗Υ Ξ2 )(ξ, η) = σ (Ξ1 ⋄γ Ξ2 )(ξ, η)
γ
⇔ Υ̂ (ξ, η)e−2⟨ξ,η⟩ = ex p[− (⟨ξ, ξ ⟩ + ⟨η, η⟩) − ⟨ξ, η⟩]
2
γ
⇔ Υ̂ (ξ, η) = ex p[− (⟨ξ, ξ ⟩ + ⟨η, η⟩) + ⟨ξ, η⟩],
2
1 ∗ 1
which is equivalent to Υ = e− 2 ∆G Γ (I )e− 2 ∆G , by using the properties (2.5) and (2.6). □

5.2. Applications: Ordinary quantum differential equation

Let γ be a complex number such that Re(γ ) ≥ 4.

Finally, as application, we resolve the following differential equation:
∂Ξ (t)
{
∂t
= At ⋄γ Ξ (t) + Bt , t ∈ [0, T ] ,
(5.3)
Ξ (0) = Ξ0 ∈ L((E), (E)∗ ) ,
where {At } and {Bt } are quantum stochastic process valued in L((E), (E)∗ ) satisfying that there
2 2
exist C ≥ 0 and p ≥ 0 such that max0≤t≤T {|σ (At )(ξ, η)|, |σ (Bt )(ξ, η)|} ≤ Ce|ξ | +|η| , for all
ξ , η ∈ E.

Theorem 5.3. The solution of the differential equation (5.3) is given by

{∫ T } { ∫ T { ∫ s } }
Ξ (t) = expγ As ds ⋄γ Ξ0 + Bs ⋄γ expγ − A y dy ds . (5.4)
0 0 0

Proof. Applying the symbol map σ to Eq. (5.3), we obtain an ordinary differential equation
of first order:
γ
∂Θ(t)
{
∂t
= e− 2 (⟨ξ,ξ ⟩+⟨η,η⟩)−⟨ξ,η⟩ σ (At )(ξ, η)Θ(t) + σ (Bt )(ξ, η) ,
Θ(0) = σ (Ξ0 )(ξ, η) ∈ (E) ,

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx 13

where Θ(t) = σ (Ξ (t))(ξ, η) for all ξ, η ∈ E. It is not difficult to see that the solution is given
by
Θ(t)
{ ∫ T }
− γ2 (⟨ξ,ξ ⟩+⟨η,η⟩)−⟨ξ,η⟩
= ex p e σ (As )(ξ, η)ds
0
{ ∫ T { ∫ s } }
− γ2 (⟨ξ,ξ ⟩+⟨η,η⟩)−⟨ξ,η⟩
× Θ(0) + σ (Bs )(ξ, η) exp e (−σ (A y )(ξ, η))dy ds .
0 0

So, applying σ −1
and using Eqs. (3.20) and (3.22), we complete the proof of the theorem. □

Example 5.1. We recall from [3] some definitions and properties.

Definition 5.1. Let Φ ∈ (E ⊗ (E))∗ . All operators ΛΦ satisfy

σ (ΛΦ )(ξ, η) = e⟨ξ,η⟩ ⟨⟨Φ, ξ ⊗ ϕγ ξ +η ⟩⟩, ∀ξ, η ∈ E, (5.5)
are called first order γ -differential operator with coefficient Φ. The operator ΛΦ has the
following integral expression:
∫
ΛΦ (ϕ) = Φ(t) ⋄γ at (ϕ)dt, ϕ ∈ (E). (5.6)
T

Examples 5.1 ([3]).

1. For example for Φ(t) = y(t) ⊗ 1 ∈ (E ⊗ (E))∗ , then ΛΦ = D y , where D y is the Gâteaux
differential operator in the direction along y ∈ E ∗ .
2. If we take Φ0 ∈ (E ⊗ (E))∗ by the formal formula
Φ0 (t, x) = x(t), t ∈ T, x ∈ E ∗ .
That is, ⟨⟨Φ0 , ξ ⊗φ⟩⟩ = ⟨ξ, f 1 ⟩ for any ξ ∈ E and φ = ( f n ) ∈ (E). Then, ΛΦ = γ ∆G +N ,
where N is the number operator.

Proposition 5.4. The solution of the differential equation (5.3) for Bt = at , is given by
{ }
Ξ (t) = Φt ⋄γ Ξ0 + ΛΦt , (5.7)
{ }
∫t
where Φt = expγ 0 As ds and ΛΦt is first order γ -differential operator with coefficient Φt .
In particular, for Φt given by
Φt (x) = x(t), t ∈ T, x ∈ E ∗ ,
then
{ }
Ξ (t) = Φt ⋄γ Ξ0 + γ ∆G + N . (5.8)

Proof. To prove Eq. (5.7) is sufficient to apply Theorem 5.3. And using the second example
given in Examples 5.1, we obtain Eq. (5.8). □

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.
14 S. Horrigue / Indagationes Mathematicae xxx (xxxx) xxx

Acknowledgments
The authors would to thank the anonymous referees for their carefully reading of this
manuscript and their useful comments and suggestions which allow us to improve our paper.

References
[1] L. Accardi, Y.-G. Lu, I. Volovich, Non-linear extensions of classical and quantum stochastic calculus and
essentially infinite dimensional analysis, in: L. Accardi, C.C. Heyde (Eds.), Probability Towards 2000, in:
Lect. Notes in Stat., vol. 128, Springer-Verlag, 1998, pp. 1–33.
[2] D.M. Chung, T.S. Chung, Wick derivations on white noise functionals, J. Korean Math. Soc. 33 (4) (1996)
993–1008.
[3] D.M. Chung, T.S. Chung, First order differential in white noise analysis, Proc. Amer. Math. Soc. 8 (1998)
2369–2376, Vol. 126.
[4] D.M. Chung, T.S. Chung, U.C. Ji, A simple proof of analytic characterization theorem for operator symbols,
Bull. Korean Math. Soc. 34 (1997) 421–436.
[5] D.M. Chung, U.C. Ji, N. Obata, Higher powers of quantum white noises in terms of integral kernel operators,
Infin. Dimens. Anal. Quantum Probab. 1 (1998) 533–559.
[6] D.M. Chung, U.C. Ji, N. Obata, Quantum stochastic analysis via white noise operators in weighted Fock
space, Rev. Math. Phys. 14 (3) (2001) 241–272.
[7] T. Hida, Analysis of Brownian Functionals, in: Carleton Math. Lect. Notes, vol. 13, Carleton University,
Ottawa, 1975.
[8] T. Hida, H.-H. Kuo, J. Potthoff, L. Streit, White Noise: An Infinite Dimensional Calculus, Kluwer Academic
Publishers, 1993.
[9] S. Horrigue, γ -Product of white noise space and applications, Random Oper. Stoch. Equ. (2017) http:
//dx.doi.org/10.1515/rose-2017-0015.
[10] Z.-Y. Huang, S.-L. Luo, Wick calculus of generalized operators and its applications to quantum stochastic
calculus, Infin. Dimens. Anal. Quantum Probab. 1 (1998) 455–466.
[11] U.C. Ji, Quantum extension of Fourier-Gauss and Fourier-Mehler transforms, J. Korean Math. Soc. 45 (6)
(2008) 1785–1801.
[12] U.C. Ji, Y.Y. Kim, Convolutions of white noise operators, Bull. Korean Math. Soc. 48 (5) (2011) 1003–1014.
[13] U.C. Ji, N. Obata, Quantum white noise calculus, in: N. Obata, T. Matsui, A. Hora (Eds.), Non-Commutativity,
Infinite-Dimensionality and Probability at the Crossroads, World Scientific, 2002, pp. 143–191.
[14] U.C. Ji, N. Obata, H. Ouerdiane, Analytic characterization of generalized Fock space operators as two-
variable entire functions with growth condition, Infin. Dimens. Anal. Quantum Probab. Relat. Top. 5 (3)
(2002) 395–407.
[15] I. Kubo, S. Takenaka, Calculus on Gaussian white noise I–IV, Proc. Japan Acad. 56A (1980) 376–380,
411–416; 57A (1981), 433–437; 58A (1982), 186–189.
[16] H.-H. Kuo, White Noise Distribution Theory, CRC Press, 1996.
[17] N. Obata, An analytic characterization of symbols of operators on white noise functionals, J. Math. Soc.
Japan 45 (1993) 421–445.
[18] N. Obata, White Noise Calculus and Fock Space, in: Lect. Notes in Math., vol. 1577, Springer, 1994.
[19] N. Obata, Integral kernel operators on Fock space generalizations and applications to quantum dynamics, Acta
Appl. Math. 47 (1997) 49–77.
[20] N. Obata, Wick product of white noise operators and quantum stochastic differential equations, J. Math. Soc.
Japan 51 (3) (1999) 613–641.

Please cite this article as: S. Horrigue, γ -Quantum product of white noise operators and applications, Indagationes Mathematicae (2020),
https://doi.org/10.1016/j.indag.2020.02.003.