The Malgrange Form and Fredholm Determinants
We consider the factorization problem of matrix symbols relative to a closed contour, i.e., a Riemann-Hilbert problem, where the symbol depends analytically on parameters. We show how to define a function τ which is locally analytic on the space of deformations and that is expressed as a Fredholm de...
Збережено в:
| Опубліковано в: : | Symmetry, Integrability and Geometry: Methods and Applications |
|---|---|
| Дата: | 2017 |
| Автор: | |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут математики НАН України
2017
|
| Онлайн доступ: | https://nasplib.isofts.kiev.ua/handle/123456789/148566 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | The Malgrange Form and Fredholm Determinants / M. Bertola // Symmetry, Integrability and Geometry: Methods and Applications. — 2017. — Т. 13. — Бібліогр.: 15 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
nasplib_isofts_kiev_ua-123456789-148566 |
|---|---|
| record_format |
dspace |
| spelling |
Bertola, M. 2019-02-18T15:57:07Z 2019-02-18T15:57:07Z 2017 The Malgrange Form and Fredholm Determinants / M. Bertola // Symmetry, Integrability and Geometry: Methods and Applications. — 2017. — Т. 13. — Бібліогр.: 15 назв. — англ. 1815-0659 2010 Mathematics Subject Classification: 35Q15; 47A53; 47A68 DOI:10.3842/SIGMA.2017.046 https://nasplib.isofts.kiev.ua/handle/123456789/148566 We consider the factorization problem of matrix symbols relative to a closed contour, i.e., a Riemann-Hilbert problem, where the symbol depends analytically on parameters. We show how to define a function τ which is locally analytic on the space of deformations and that is expressed as a Fredholm determinant of an operator of ''integrable'' type in the sense of Its-Izergin-Korepin-Slavnov. The construction is not unique and the non-uniqueness highlights the fact that the tau function is really the section of a line bundle. The author wishes to thank Oleg Lisovyy for asking a very pertinent question on the representation of the Malgrange form in terms of Fredholm determinants. Part of the thinking was done during the author’s stay at the “Centro di Ricerca Matematica Ennio de Giorgi” at the Scuola Normale Superiore in Pisa, workshop on “Asymptotic and computational aspects of complex dif ferential equations” organized by G. Filipuk, D. Guzzetti and S. Michalik. The author wishes to thank the organizers and the Institute for providing an opportunity of fruitful exchange. en Інститут математики НАН України Symmetry, Integrability and Geometry: Methods and Applications The Malgrange Form and Fredholm Determinants Article published earlier |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| title |
The Malgrange Form and Fredholm Determinants |
| spellingShingle |
The Malgrange Form and Fredholm Determinants Bertola, M. |
| title_short |
The Malgrange Form and Fredholm Determinants |
| title_full |
The Malgrange Form and Fredholm Determinants |
| title_fullStr |
The Malgrange Form and Fredholm Determinants |
| title_full_unstemmed |
The Malgrange Form and Fredholm Determinants |
| title_sort |
malgrange form and fredholm determinants |
| author |
Bertola, M. |
| author_facet |
Bertola, M. |
| publishDate |
2017 |
| language |
English |
| container_title |
Symmetry, Integrability and Geometry: Methods and Applications |
| publisher |
Інститут математики НАН України |
| format |
Article |
| description |
We consider the factorization problem of matrix symbols relative to a closed contour, i.e., a Riemann-Hilbert problem, where the symbol depends analytically on parameters. We show how to define a function τ which is locally analytic on the space of deformations and that is expressed as a Fredholm determinant of an operator of ''integrable'' type in the sense of Its-Izergin-Korepin-Slavnov. The construction is not unique and the non-uniqueness highlights the fact that the tau function is really the section of a line bundle.
|
| issn |
1815-0659 |
| url |
https://nasplib.isofts.kiev.ua/handle/123456789/148566 |
| citation_txt |
The Malgrange Form and Fredholm Determinants / M. Bertola // Symmetry, Integrability and Geometry: Methods and Applications. — 2017. — Т. 13. — Бібліогр.: 15 назв. — англ. |
| work_keys_str_mv |
AT bertolam themalgrangeformandfredholmdeterminants AT bertolam malgrangeformandfredholmdeterminants |
| first_indexed |
2025-11-25T07:15:11Z |
| last_indexed |
2025-11-25T07:15:11Z |
| _version_ |
1850510054795509760 |
| fulltext |
Symmetry, Integrability and Geometry: Methods and Applications SIGMA 13 (2017), 046, 12 pages
The Malgrange Form and Fredholm Determinants
Marco BERTOLA †‡
† Department of Mathematics and Statistics, Concordia University, Montréal, Canada
E-mail: marco.bertola@concordia.ca
‡ Area of Mathematics SISSA/ISAS, Trieste, Italy
E-mail: marco.bertola@sissa.it
Received March 12, 2017, in final form June 17, 2017; Published online June 22, 2017
https://doi.org/10.3842/SIGMA.2017.046
Abstract. We consider the factorization problem of matrix symbols relative to a closed
contour, i.e., a Riemann–Hilbert problem, where the symbol depends analytically on pa-
rameters. We show how to define a function τ which is locally analytic on the space of
deformations and that is expressed as a Fredholm determinant of an operator of “inte-
grable” type in the sense of Its–Izergin–Korepin–Slavnov. The construction is not unique
and the non-uniqueness highlights the fact that the tau function is really the section of a line
bundle.
Key words: Malgrange form; Fredholm determinants; tau function
2010 Mathematics Subject Classification: 35Q15; 47A53; 47A68
1 Introduction
We shall consider the following prototypical matrix Riemann–Hilbert problem (RHP) on the
unit circle Σ (or any smooth closed simple contour):
Γ+(z; t) = Γ−(z; t)M(z; t), ∀ z ∈ Σ, Γ(∞) = 1. (1.1)
Here t stands for a vector of parameters which we refer to as “deformation parameters”. The
assumptions are the following;
1. The matrix M(z; t) ∈ GLn(C) is jointly analytic for z in a fixed tubular neighbour-
hood N(Σ) of Σ and t in an open connected domain S, which we refer to as the “defor-
mation space”.
2. The index of detM(z; t) around Σ vanishes for all t ∈ S.
3. The partial indices are generically zero, i.e., the RHP (1.1) generically admits solution.
Let us remind the reader of some facts that can be extracted from [7]
– There exists a matrix function Y−(z) analytic and analytically invertible in Ext(Σ)∪N(Σ)
(and uniformly bounded) and similarly a matrix function Y+(z) analytic and analytically
invertible in Int(Σ) ∪N(Σ) and n integers k1, . . . , kn (called partial indices) such that
D(z)Y+(z) = Y−(z)M(z), z ∈ Σ, D = diag
(
zk1 , . . . , zkn
)
.
– The RHP (1.1) is solvable if and only if all partial indices vanish, kj = 0, ∀ j = 1, . . . , n.
Note that since indΣ detM =
n∑
j=1
kj , the condition (2) in our assumptions is necessary for the
solvability of (1.1).
mailto:marco.bertola@concordia.ca
mailto:marco.bertola@sissa.it
https://doi.org/10.3842/SIGMA.2017.046
2 M. Bertola
We shall denote by D± the interior (+) and the exterior (−) regions separated by Σ. Defi-
ne H+ to be the space of functions that are in L2(Σ, |dz|) and extend to analytic functions
in the interior. We will use the notation ~H+ = H+ ⊗ Cr (i.e., vector-valued such func-
tions). The vectors will be thought of as row-vectors. We also introduce the Cauchy projectors
C± : L2(Σ, |dz|)→ H±:
C±[f ](z) =
∮
Σ
f(w)dw
(w − z)2iπ
, z ∈ D±.
It is well known [14] that the RHP (1.1) is solvable if and only if the Toeplitz operator1
TS : ~H+ → ~H+,
TS [~f ] = C+[~fS], S(z; t) := M−1(z; t)
is invertible, in which case the inverse is given by
T−1
S [~f ] = C+
[
~fΓ−1
−
]
Γ+.
Moreover the operator is Fredholm and
dim ker(TS)− dim coker(TS) = indΣ detM = 0.
There is no reasonable way, however, to define a “determinant” of TS as it stands. Such
a function of t would desirably have the property that the RHP (1.1) is not solvable if and only
if this putative determinant is zero.
While this is notoriously impossible in this naive form, we now propose a proxy for the notion
of determinant, in terms of a simple Fredholm determinant.
The Malgrange one-form. As Malgrange explains [15] one can define a central extension
on the loop group G := {M : Σ → GLn(C) : indΣ detM = 0} given by Ĝ = {(M,u) ∈ G × C×}
with the group law2
(M,u) · (M̃, ũ) =
(
MM̃, uũc
(
M, M̃
))
, c(M, M̃) := det
H+
(
TM−1T
M̃−1T
−1
(MM̃)−1
)
.
The operator in the determinant is of the form IdH+ + (trace class) and hence the Fredholm
determinant is well defined. This group law is only valid for pairs M , M̃ for which the inverse
of T−1
(MM̃)−1
exists. The left-invariant Maurer–Cartan form of this central extension is then given
by
(
S−1δS, du
u + ω̂M
)
, where
ω̂M := TrH+
(
T−1
S ◦ TδS − TS−1δS
)
, S := M−1,
and this can be written as the following integral
ω̂M =
∮
Σ
Tr
(
Γ−1
+ Γ′+M
−1δM
) dz
2iπ
. (1.2)
Here, and below, δ denotes the exterior total differentiation in the deformation space S:
δ =
∑
δtj
∂
∂tj
.
1Due to our choices of symbols, the matrix symbol of the relevant Toeplitz operator is M−1. We apologize for
the inconvenience.
2We are being a bit cavalier in this description; we invite the reader to read pp. 1373–1374 in [15].
The Malgrange Form and Fredholm Determinants 3
The Malgrange form is a logarithmic form in the sense that it has only simple poles on
a co-dimension 1 analytic submanifold of the deformation space S and with positive integer
Poincaré residue along it; this manifold is precisely the exceptional “divisor” (Θ) ⊂ S (the
Malgrange divisor) where the RHP (1.1) becomes non solvable, i.e., where some partial indices
of the Birkhoff factorization become non-zero.
Closely related to (1.2) is the following one-form, which we still name after Malgrange:
ωM :=
∮
Σ
Tr
(
Γ−1
− Γ′−δMM−1
) dz
2iπ
. (1.3)
It is also a logarithmic form with the same pole-divisor; indeed one verifies that
ω̂M − ωM =
∮
Σ
Tr
(
M ′M−1δMM−1
) dz
2iπ
,
which is an analytic form of the deformation parameters t ∈ S. In [2] the one-form (1.3) was
posited as an object of interest for general Riemann–Hilbert problems (not necessarily on closed
contours) and its exterior derivative computed (with an important correction in [3], which is
however irrelevant in the present context). It was computed (but the computation can be traced
back to Malgrange himself in this case) that
δωM =
1
2
∮
Σ
Tr
(
Ξ(z) ∧ d
dz
Ξ(z)
)
dz
2iπ
, Ξ(z; t) := δM(z; t)M(z; t)−1. (1.4)
It appears from (1.4) that this two form δωM is not only closed, but also smooth on the whole
of S, including the Malgrange-divisor. As such, it defines a line-bundle L over S by the usual
construction: one covers S by appropriate open sets Uα where δωM = δθα; on the overlap
Uα∩Uβ the form θα− θβ is also exact and one defines then the transition functions by gαβ(t) =
exp
( ∫
θα − θβ
)
. Then a section of this line bundle is provided by the collection of functions
τα : Uα → C such that
τα(t) = exp
[∫
(ωM − θα)
]
.
Since ωM is a logarithmic form and each θα is analytic in the respective Uα, the functions τα(t)
have zero of finite order precisely on the Malgrange divisor (Θ) ⊂ S (under appropriate transver-
sality assumptions, the order of the zero is the dimension of KerTS).
Our goal is to provide an explicit construction of the τα’s in terms of Fredholm determinants
of simple operators of the Its–Izergin–Korepin–Slavnov “integrable” type [13]. Their definition
is recalled in due time.
2 Construction of the Fredholm determinants
The construction carried out below is not unique, and also only local in the deformation space S;
this is however not only not a problem, but rather an interesting feature, as we will illustrate in
the case of SL2(C). The non-uniqueness is precisely a consequence of the fact that we are trying
to compute a section of the aforementioned line bundle.
Preparatory step. The assumption that M(z; t) ∈ GLn can be replaced without loss of
generality with M(z; t) ∈ SLn; this is so because of the assumption on the index of detM .
Indeed we can solve the scalar problem y+(z) = y−(z) detM(z), y(∞) = 1 and then define
a new RHP where Γ̃±(z) := Γ± diag
(
y−1
± , 1, . . .
)
and hence the new matrix jump for Γ̃ is
M̃(z) = diag(y−(z), 1, . . . )M(z) diag
(
y−1
+ (z), 1, . . .
)
with det M̃ ≡ 1. For this reason, from here
on we assume M ∈ SLn(C).
We define an elementary matrix (for our purposes) to be a matrix of the form 1 + cEjk, with
j 6= k, where Ejk denotes the (j, k)-unit matrix.
4 M. Bertola
Lemma 2.1. Any matrix M ∈ SLn(C) can be written as a product of elementary matrices.
The entries of the factorization are rational in the entries of M with denominators that are
monomials in a suitable set of n− 1 nested minors of M .
Proof. We recall that given any matrix M ∈ SLn, there is a permutation Π of the columns such
that the principal minors (the determinants of the top left square submatrices) do not vanish,
and hence we can write it as
M = LDUΠ = M̂Π,
where L, U are lower/upper triangular matrices with unit on the diagonal and D = diag(x1, . . . ,
xn) is a diagonal matrix (see for example [8, Vol. 1, Chapter II]). Denote q` = det
[
M̂j,k
]
j,k≤`
the principal minors; these are the nested minors of the original matrix M alluded to in the
statement. The matrices L, U are rational in the entries of M and with denominators that are
monomials in the q`’s.
Now, both L, U can clearly be written as products of elementary matrices whose coefficients
are polynomials in the entries of L, U (respectively) and so it remains to show that we can
write D as product of elementary matrices.
To this end we observe the ’LULU’ identity (there is a similar ‘ULUL’ identity)[
x 0
0 1
x
]
=
[
1 0
1−x
x 1
] [
1 1
0 1
] [
1 0
x− 1 1
] [
1 − 1
x
0 1
]
.
As D ∈ SLn, we can represent it in terms of product of embedded SL2 matrices using the root
decomposition of SLn:
D = diag
(
x1,
1
x1
, 1, . . . , 1
)
diag
(
1, x2x1,
1
x1x2
, 1, . . . , 1
)
× diag
(
1, 1, x1x2x3,
1
x1x2x3
, 1, . . . , 1
)
· · · ,
and then embed the LULU identity for each factor.
Finally, also permutation matrices can be written as product of elementary matrices embed-
ding appropriately the simple identity;[
0 −1
1 0
]
=
[
1 0
1 1
] [
1 −1
0 1
] [
1 0
1 1
]
.
This concludes the proof. �
Let now M(z) be an SLn matrix valued function, analytic in a tubular neighbourhood N(Σ)
of Σ, and let q`(z), ` = 1, . . . , n − 1 be the nested minors alluded to in the Lemma so that
they are not identically zero. Since the entries are analytic in N(Σ) we can slightly deform the
contour Σ to a contour Σ̃ that avoids all zeroes of every principal minor q`(z). The resulting
RHP is “equivalent” to the original in the sense that the solvability of one implies the solvability
of the other. Note also that this deformation can be done in a piecewise constant way locally
with respect to t ∈ S. Thus we have
M(z) = F1(z) · · ·FR(z), Fν(z) = 1 + aν(z)Ejν ,kν , ν = 1, . . . , R.
Corresponding to this factorization we can define an equivalent RHP with jumps on R con-
tours Σ1, . . . ,ΣR, with Σ1 = Σ and Σj+1 in the interior of Σj and all of them in the joint domain
of analyticity of the scalar functions aν(z) (see Fig. 1) which may have poles only at the zeroes
The Malgrange Form and Fredholm Determinants 5
Σ = Σ1
D
1
Σ2
D
2
Σ3 D
R−1
ΣR
DR = D+
D0 = D−
Figure 1. An illustration of the splitting of the jump matrix into elementary jumps. The matrix M(z; t)
is analytic in the shaded regions.
of the principal minors q`(z) of M(z). This is accomplished by “extending” the matrix Γ−(z)
to the annular regions D0 = D− and Dj = Int(Σj) ∩ Ext(Σj+1) as
Θ0(z) := Γ−(z), ∀ z ∈ D0,
Θν(z) := Γ−(z)
−→
ν∏
`=1
F`(z), ∀ z ∈ Dν . (2.1)
By doing so we obtain the following relations
Θν(z) = Θν−1(z)Fν(z), ∀ z ∈ Σν , ν = 1, . . . , R.
The piecewise analytic matrix function Θ(z) whose restriction to Dν coincides with the matri-
ces Θν (2.1), satisfies a final RHP
Θ+(z) = Θ−(z)Fν(z), ∀ z ∈ Σν , Θ(∞) = 1. (2.2)
This type of RHP is of the general type of “integrable kernels” and its solvability can be
determined by computing the Fredholm determinant of an integral operator of L2(
⊔
Σν , |dz|) '⊕R
ν=1 L
2(Σν , |dz|) with kernel (we use the same symbol for the operator and its kernel)
K(z, w) =
~fT (z)~g(w)
2iπ(w − z)
, ~f(z) =
R∑
ν=1
ejνχν(z)aν(z), ~g(z) =
R∑
ν=1
ekνχν(z), (2.3)
where χν(z) is the projector (indicator function) on the component L2(Σν , |dz|). Indeed, as
explained in [4, 10, 13], the Fredholm determinant det(Id−K) is zero if and only if the RHP (2.2)
is non-solvable and moreover the resolvent operator R = K(Id−K)−1 of K has kernel
R(z, w) =
~f t(z)Θt(z)(Θt)−1(w)~g(w)
z − w
.
Theorem 2.2. The RHP (2.2) and hence (1.1) is solvable if and only if τ := det(Id−K) 6= 0.
Proposition 2.3 (see, e.g., [4, Theorem 2.1]). Let ∂ be any deformation of the functions aν(z),
then
∂ ln τ =
R∑
ν=1
∮
Σν
Tr
(
Θ−1
− Θ′−(z)∂FνF
−1
ν
) dz
2iπ
=
R∑
ν=1
∮
Σν
(
Θ−1
ν−1Θ′ν−1(z)
)
kν ,jν
∂aν(z)
dz
2iπ
.
6 M. Bertola
3 The SL2 case
We would like to express the Malgrange one-form directly in terms of the τ function (Fredholm
determinant). Rather than obscuring the simple idea with the general case, we consider in detail
the SL2 case. Let M(z; t) be analytic in (z, t) ∈ N(Σ) × S and with values in SL2(C). Using
the general scheme above, we have the following factorizations
(1)
[
a b
c d
]
=
[
1 0
1+c−a
a 1
]
︸ ︷︷ ︸
F1(z)
[
1 1
0 1
]
︸ ︷︷ ︸
F2(z)
[
1 0
a− 1 1
]
︸ ︷︷ ︸
F3(z)
[
1 b−1
a
0 1
]
︸ ︷︷ ︸
F4(z)
, a 6≡ 0,
(2)
[
a b
c d
]
=
[
1 1+b−d
d
0 1
]
︸ ︷︷ ︸
F1(z)
[
1 0
1 1
]
︸ ︷︷ ︸
F2(z)
[
1 d− 1
0 1
]
︸ ︷︷ ︸
F3(z)
[
1 0
c−1
d 1
]
︸ ︷︷ ︸
F4(z)
, d 6≡ 0,
(3)
[
a b
−1
b 0
]
=
[
1 b− ab
0 1
]
︸ ︷︷ ︸
F1(z)
[
1 0
−1
b 1
]
︸ ︷︷ ︸
F2(z)
[
1 b
0 1
]
︸ ︷︷ ︸
F3(z)
,
(4)
[
0 b
−1
b d
]
=
[
1 d−1
b
0 1
]
︸ ︷︷ ︸
F1(z)
[
1 0
b 1
]
︸ ︷︷ ︸
F2(z)
[
1 −1
b
0 1
]
︸ ︷︷ ︸
F3(z)
,
(5)
[
0 b
−1
b 0
]
=
[
1 0
b 1
]
︸ ︷︷ ︸
F1(z)
[
1 −1
b
0 1
]
︸ ︷︷ ︸
F2(z)
[
1 0
b 1
]
︸ ︷︷ ︸
F3(z)
. (3.1)
3.1 Fredholm determinants for different factorizations
Each of the factorization (3.1) leads to an integrable operator of the form (2.3) and hence to
a corresponding Fredholm determinant; we now establish their mutual relationships.
There are two types of questions that we address here
1. How are the Fredholm determinants associated with the different factorizations (1, 2)
in (3.1) related to each other?
2. For a fixed factorization, how does the Fredholm determinant depend on the choice of
contour Σ (within the analyticity domain N(Σ)).
Consider the cases (3.1)(ρ), ρ = 1, . . . , 5. We compute the logarithmic derivative of the
corresponding Fredholm determinant using Proposition 2.3
∂ ln τ(ρ) =
R∑
ν=1
∮
Σν
Tr
(
Θ−1
− Θ′−∂FνF
−1
ν
) dz
2iπ
. (3.2)
By using the relationship (2.1) between Γ− (extended to an analytic function on Ext(Σ)∪N(Σ)),
we can re-express it in terms of the Malgrange one-form of the original problem (1.1); we use
the fact that we can deform the contours back to Σ = Σ1 by Cauchy’s theorem. Plugging (2.1)
appropriately in (3.2) and using Leibnitz rule, after a short computation we obtain
δ ln τ
(ρ)
=
∮
Σ
Tr
(
Γ−1
− Γ′−∂MM−1
) dz
2iπ
(3.3)
+
∮
Σ
Tr
(
F−1
1 F ′1∂F2F
−1
2 + F−1
12 F
′
12∂F3F
−1
3 + F−1
123F
′
123∂F4F
−1
4
) dz
2iπ
=: ωM + θ(ρ),
where F1...k = F1F2F3 · · ·Fk (if the factorization has only three term, then we set F4 ≡ 1).
The Malgrange Form and Fredholm Determinants 7
The cases (3,4,5). The last cases (3.1)(3,4,5) lead essentially to a RHP with a triangular
jump; it suffices to re-define Γ by Γ
[
0 1
−1 0
]
for z ∈ Int(Σ). If the index of b is zero, indΣ b = 0,then
the solution can be written explicitly in closed form and it is interesting to compute the Fredholm
determinant associated to our factorization of the matrix. We will show
Proposition 3.1. In the cases (3.1)(3,4,5) and under the additional assumption that indΣ b = 0,
the τ function given by det(IdL2(∪Σj)−K) and K as in (2.3), equals the constant in the strong
Szegö formula, given by [1, 5]
τ = exp
[∑
j>0
jβ−jβj
]
= det
H+
TbTb−1 ,
where now the Toeplitz operator is for the scalar symbol b(z) and H+ is the Hardy space of scalar
functions analytic in D+ and βj are the coefficients of ln b(z) in the Laurent expansion centered
at the origin
β(z) := ln b(z) =
∑
j∈Z
βjz
j . (3.4)
The same applies in the case that the jump is triangular (b ≡ 0 and/or c ≡ 0) under the
assumption indΣ a = 0 and replacing b with a in the above formulas.
Proof. From a direct computation of the term θ(ρ) in (3.3), we find
θ
(3,4,5)
=
∮
Σ
(
δβ
d
dz
β
)
dz
2iπ
, β(z) := ln b(z)
and the solution of the RHP is explicit,
Γ(z) =
eB(z) −e−B(z)
∮
Σ
a(w)b(w)e2B−(w)
w − z
dw
2iπ
0 e−B(z)
, z ∈ D−,eB(z) −e−B(z)
∮
Σ
a(w)b(w)e2B−(w)
w − z
dw
2iπ
0 e−B(z)
[ 0 1
−1 0
]
, z ∈ D+,
B(z) :=
∮
Σ
β(w)
w − z
dw
2iπ
.
Thus we can write explicitly the Malgrange form:
ωM =
∮
Σ
Tr
(
Γ−1
− Γ′−δMM−1
) dz
2iπ
=
∮
Σ
Tr
(
σ3
∮
Σ
β(w)
(w − z−)2
dw
2iπ
δβ(z)σ3
)
dz
2iπ
= 2
∮
Σ−
∮
Σ
β(w)δ(β(z))
(w − z)2
dw
2iπ
dz
2iπ
.
In this integral, z is integrated on a slightly “larger” contour Σ−. The Fredholm determinant
satisfies
δ ln τ(3,4,5) = ωM + θ
(3,4,5)
= 2
∮
Σ−
∮
Σ
β(w)δ(β(z))
(w − z)2
dw
2iπ
dz
2iπ
+
∮
Σ
[
β(δβ)′
] dw
2iπ
.
We now show that
δ ln τ = δ
∮
Σ−
∮
Σ
β(w)β(z)
(w − z)2
dw
2iπ
dz
2iπ
=
∮
Σ−
∮
Σ
δβ(w)β(z) + β(w)δβ(z)
(w − z)2
dw
2iπ
dz
2iπ
.
8 M. Bertola
Indeed, the exchange of order of integration of one of the addenda (and relabeling the variables)
yields the other term plus the residue on the diagonal,∮
Σ−
∮
Σ
δβ(w)β(z) + β(w)δβ(z)
(w − z)2
dw
2iπ
dz
2iπ
=
∮
Σ−
∮
Σ
2β(w)δβ(z)
(w − z)2
dw
2iπ
dz
2iπ
+
∮
Σ
β(z)(δβ)′
dz
2iπ
.
Therefore, in conclusion, we have (we fix the overall constant of τ by requiring it to be 1 for
b ≡ 1)
ln τ =
∮
Σ−
∮
Σ
ln b(w) ln b(z)
(w − z)2
dw
2iπ
dz
2iπ
. (3.5)
If we write a Laurent expansion of ln b (3.4) the formula (3.5) gives the explicit expression
ln τ =
∑
j>0
jβ−jβj ,
which is also the formula for the second Szegö limit theorem for the limit of the Toeplitz determi-
nants of the symbol b(z), and it is known to to be the Fredholm determinant of an operator [1, 5]
τ = det
H+
TbTb−1 ,
where now the Toeplitz operator is for the scalar symbol b(z) and H+ is the Hardy space of
scalar functions analytic in D+. �
Remark 3.2. The index assumption is only necessary for the case (3.1)(5) or (3.1)(1) when
b ≡ 0 ≡ c, because in these situations the RHP separates into two scalar problems. However,
the assumption indΣ b = 0 for cases (3,4) or indΣ a = 0 for the triangular case is not necessary
as we now show (in the latter form). Consider the RHP
Y+ = Y−
[
1 µ(z)
0 1
]
, z ∈ Σ, Y (z) =
(
1 +O
(
z−1
))
znσ3 , z →∞.
By defining
Γ(z) =
{
Y (z), z ∈ Int(Σ),
Y (z)z−nσ3 , z ∈ Ext(Σ),
we are lead to the RHP in standard form
Γ+ = Γ−
[
zn znµ(z)
0 z−n
]
, z ∈ Σ, Γ(∞) = 1,
which is case (1) with c ≡ 0 (or essentially cases (3,4) up to a multiplication by piecewise constant
matrices). Even if a(z) = zn and thus indΣ a = n, this problem is still generically solvable
in terms of appropriate “orthogonal polynomials” {p`(z)}`∈N defined by the “orthogonality”
property∮
Σ
p`(z)pk(z)µ(z)dz = h`δ`k.
The Malgrange Form and Fredholm Determinants 9
Then the solution of the Y -problem is written as
Y (z) =
pn(z)
∮
Σ
pn(w)µ(w)dw
(w − z)2iπ
−2iπpn−1(z)
hn−1
∮
Σ
−pn−1(w)µ(w)dw
(w − z)hn−1
,
and the solvability depends only on the condition [6]
det
[∮
Σ
z`+j−2µ(z)dz
]n
j,`=1
6= 0.
The cases (1,2). A straightforward computation using the explicit expression (3.1)(1) yields
∂ ln τ
(1)
=
∮
Σ
[
Tr
(
Γ−1
− Γ′−∂MM−1
)
+ ∂ ln a(ln a)′(1 + bc)
− c∂b(ln a)′ + c′∂b− c′b∂(ln a)
] dz
2iπ
=
∮
Σ
Tr
(
Γ−1Γ′∂MM−1
) dz
2iπ
+
∮
Σ
(
(a)′
a
(d∂a− c∂b) + ac′∂
(
b
a
))
dz
2iπ︸ ︷︷ ︸
θ
(1)
. (3.6)
Since ∂ ln τ is a closed differential, the exterior derivative of θ
(1)
must be opposite to the one of
the first term, which is given by (1.4). Let us verify this directly; to this end we compute the
exterior derivative of θ
(1)
. A straightforward computation yields
δθ
(1)
=
∮
Σ
(
δb ∧ (δc)′ +
δa ∧ (δa)′
a2
(1 + bc) + δa ∧
(
δc
a
)′
b
− δa ∧ δb(c)′
a
− δb ∧ (δa)′
c
a
− δb ∧ δca
′
a
)
dz
2iπ
.
The exterior derivative of ωM is given by (1.4), in which we can insert the explicit expression
of M(z; t); after a somewhat lengthy but straightforward computation we find that
δωM + δθ
(1)
= −1
2
∮
Σ
d
dz
(
c
a
δa ∧ δb− b
a
δa ∧ δc+ δb ∧ δc
)
dz
2iπ
= 0,
thus confirming that the differential δ ln τ
(1)
is indeed closed (of course it must be, since the tau
function is a Fredholm determinant!). The case (2) is analogous with the replacements b ↔ c
and a↔ d.
3.1.1 Determinants for different choices of Σ
The factorizations (3.1)(1,2) require that we deform the contour Σ so that a(z) (or d(z)) does
not have any zero on Σ. This leads to completely equivalent RHPs of the form (1.1) but not
entirely equivalent RHP when expressed in the form (2.2).
Note that (3.1)(3,4,5) do not suffer from this ambiguity, because b(z) cannot have any zeroes
in N(Σ) since we assumed analyticity of the jump matrix M(z; t).
Consider the factorization (3.1)(1) (with similar considerations applying to the other factor-
ization); in general a(z; t) has zeroes in its domain of analyticity, and their positions depend
on t. Therefore it may be necessary, when considering the dependence on t, to move the contour
so that certain zeroes are to the left or to the right of it because a zero may sweep across N(Σ)
as we vary t ∈ S.
10 M. Bertola
Σ̃
Σ
Figure 2. The two contours and the zeroes of a(z; t) within the enclosed region. The shaded area is
N(Σ) where M(z; t) is analytic (uniformly w.r.t. t ∈ S).
So, let Σ, Σ̃ be two contours in the common domain of analyticity of M(z; t) and such
that a(z; t) has no zeroes on either one and Σ ⊂ Int(Σ̃) (see Fig. 2).
Denote with τ
(1)
and τ̃
(1)
the corresponding Fredholm determinants of the operators defined
as described; following the same steps as above. Our goal is to show that
Theorem 3.3. The ratio of the two Fredholm determinants is given by
τ̃
(1)
= τ
(1)
∏
v∈Int(Σ̃)∩Ext(Σ):
a(v)=0
(c(v(t); t))−ordv(a).
Note that the evaluation of c at the zeroes of a cannot vanish because detM ≡ 1.
Proof. From the formula (3.6) we get
∂ ln
τ̃
(1)
τ
(1)
=
(∮
Σ̃
−
∮
Σ
)
Tr
(
Γ−1
− Γ′−∂MM−1
) dz
2iπ
+
(∮
Σ̃
−
∮
Σ
)(
(a)′
a
(d∂a− c∂b) + ac′∂
(
b
a
))
dz
2iπ
.
Here Γ− means the analytic extension of the solution Γ to the region Ext(Σ) ∪ N(Σ). Since
the integrand of the first term is holomorphic in the region bounded by Σ̃ and Σ, it yields
a zero contribution by the Cauchy’s theorem and we are left only with the second term, which
is computable by the residue theorem;
∂ ln
τ̃
(1)
τ
(1)
=
∑
v∈Int(Σ̃)∩Ext(Σ):
a(v)=0
res
z=v
(
(a)′
a
(d∂a− c∂b) + c′a∂
(
b
a
))
.
We are assuming that M(z; t) is analytic and also that detM ≡ 1. Now note that a zero v(t)
of a(z; t) in general depends on t; suppose that a(z; t) = (z − v(t))k(Ck(t) +O(z − v(t))); then
∂a(z; t)
a(z; t)
= − k∂v(t)
z − v(t)
+O(1),
and hence the residue evaluation at z = v(t) yields (we use 0 = a′∂v + ∂a|v and ∂(ad)|v =
d∂a|v = ∂(bc)|v)
res
z=v
(
(a)′
a
(d∂a− c∂b) + ac′∂
(
b
a
))
= k
(
−da′
∣∣
v
∂v − c∂b
∣∣
v
)
+ kbc′∂v
= k
(
d∂a
∣∣
v
− c∂b
∣∣
v
)
+ kbc′∂v = k
(
∂(bc)
∣∣
v
− c∂b
∣∣
v
)
+ kbc′∂v
= kb∂c
∣∣
v
+ kbc′
∣∣
v
∂v
bc|v=−1
= −k∂ ln c(v(t); t)
(note that c(v(t); t) cannot be zero because detM ≡ 1 and z = v is already a zero of a). �
The Malgrange Form and Fredholm Determinants 11
3.1.2 Determinants for different factorizations
The tau functions (Fredholm determinants) defined thus far should be understood as defining
a section of a line bundle over the loop group space; this is the line bundle associated to the two
fom δωM (1.4).
This simply means that on the intersection of the open sets where the factorizations (1), (2)
in (3.1) can be made, we have
δ ln
τ
(1)
τ
(2)
= δ ln
(
Υ(1,2)
)
,
where
δ ln
(
Υ(1,2)
)
:= δ
(
θ
(1)
− θ
(2)
)
.
After a short computation we obtain
δ ln
(
Υ(1,2)
)
=
∮
Σ
(
c′δb− b′δc
1 + bc
)
dz
2iπ
=
∮
Σ
(
δ ln b
d
dz
ln(ad)− δ ln(ad)
d
dz
ln b
)
dz
2iπ
.
Observe the last expression; in principle the functions b(z), a(z)d(z) may have nonzero index
around Σ, but in any case the functions δ ln b and (ln(ad))′ are single-valued because the incre-
ments of the logarithms are integer multiples of 2iπ and hence locally constant in the space of
deformations. To write explicitly the transition function (or rather a representative of the same
cocycle class) we need to choose a point z0 ∈ Int(Σ); we choose z = 0 without loss of generality.
Let K = indΣ b, L = indΣ(ad); then we can rewrite the above expression as follows
δ ln
(
Υ(1,2)
)
=
∮
Σ
[
δ ln
(
b
zK
)
d
dz
ln
(
ad
zL
)
− δ ln
(
ad
zL
)
d
dz
ln
(
b
zK
)
− δ ln
(
ad
zL
)
K
z
+ δ ln
(
b
zK
)
L
z
]
dz
2iπ
,
and after integration by parts (which is now possible since all functions involved are single-valued
in N(Σ))
δ ln
(
Υ(1,2)
)
= δ
∮
Σ
[
ln
(
b
zK
)
d
dz
ln
(
ad
zL
)
− ln
(
ad
zL
)
K
z
+ ln
(
b
zK
)
L
z
]
dz
2iπ
,
so that the transition function admits the explicit expression
Υ(1,2) = exp
[∮
Σ
[
ln
(
b
zK
)
d
dz
ln
(
ad
zL
)
− ln
(
ad
zL
)
K
z
+ ln
(
b
zK
)
L
z
]
dz
2iπ
]
.
4 Conclusion
We conclude this short note with a few comments.
First of all the choice of Σ as a single closed contour is not necessary; we can have a disjoint
union of closed contours or an unbounded contour as long as M(z) converges to the identity
sufficiently fast as |z| → ∞ or near the endpoints. The considerations extend with trivial
modifications. Our approach is similar in spirit to the approach used in [9] to express the
tau function of a general isomonodromic system with Fuchsian singularities in terms of an
appropriate Fredholm determinant.
Much less clear to the writer is how to handle the case where Σ contains intersections;
in this case we should stipulate a local “no-monodromy” condition at the intersection points
12 M. Bertola
as explained in [2, 3]. The obstacle is not the issue of factorization but the fact that the
resulting RHP of the IIKS type leads to an operator K (2.3) which is not of trace-class (and
not even Hilbert–Schmidt, which would be sufficient in order to construct a Hilbert–Carleman
determinant).
Nonetheless, the two form δωM defines a line bundle as explained and therefore it is possible
to compute the “transition functions” of the line-bundle; this is precisely what is accomplished
(in different setting and in different terminology) in recent works [11, 12] and the transition
functions can be expressed in terms of explicit expressions, analogously to what we have shown
here in this general but simplified setting.
Acknowledgements
The author wishes to thank Oleg Lisovyy for asking a very pertinent question on the represen-
tation of the Malgrange form in terms of Fredholm determinants. Part of the thinking was done
during the author’s stay at the “Centro di Ricerca Matematica Ennio de Giorgi” at the Scuola
Normale Superiore in Pisa, workshop on “Asymptotic and computational aspects of complex
differential equations” organized by G. Filipuk, D. Guzzetti and S. Michalik. The author wishes
to thank the organizers and the Institute for providing an opportunity of fruitful exchange.
References
[1] Basor E.L., Widom H., On a Toeplitz determinant identity of Borodin and Okounkov, Integral Equations
Operator Theory 37 (2000), 397–401, math.FA/9909010.
[2] Bertola M., The dependence on the monodromy data of the isomonodromic tau function, Comm. Math.
Phys. 294 (2010), 539–579, arXiv:0902.4716.
[3] Bertola M., Corrigendum: The dependence on the monodromy data of the isomonodromic tau function,
arXiv:1601.04790.
[4] Bertola M., Cafasso M., The transition between the gap probabilities from the Pearcey to the Airy process –
a Riemann–Hilbert approach, Int. Math. Res. Not. 2012 (2012), 1519–1568, arXiv:1005.4083.
[5] Borodin A., Okounkov A., A Fredholm determinant formula for Toeplitz determinants, Integral Equations
Operator Theory 37 (2000), 386–396, math.CA/9907165.
[6] Chihara T.S., An introduction to orthogonal polynomials, Mathematics and its Applications, Vol. 13, Gordon
and Breach Science Publishers, New York – London – Paris, 1978.
[7] Clancey K.F., Gohberg I., Factorization of matrix functions and singular integral operators, Operator Theory:
Advances and Applications, Vol. 3, Birkhäuser Verlag, Basel – Boston, Mass., 1981.
[8] Gantmacher F.R., The theory of matrices. Vols. 1, 2, Chelsea Publishing Co., New York, 1959.
[9] Gavrylenko P., Lisovyy O., Fredholm determinant and Nekrasov sum representations of isomonodromic tau
functions, arXiv:1608.00958.
[10] Harnad J., Its A.R., Integrable Fredholm operators and dual isomonodromic deformations, Comm. Math.
Phys. 226 (2002), 497–530, solv-int/9706002.
[11] Its A., Lisovyy O., Prokhorov A., Monodromy dependence and connection formulae for isomonodromic tau
functions, arXiv:1604.03082.
[12] Its A., Lisovyy O., Tykhyy Yu., Connection problem for the sine-Gordon/Painlevé III tau function and
irregular conformal blocks, Int. Math. Res. Not. 2015 (2015), 8903–8924, arXiv:1403.1235.
[13] Its A.R., Izergin A.G., Korepin V.E., Slavnov N.A., Differential equations for quantum correlation functions,
Internat. J. Modern Phys. B 4 (1990), 1003–1037.
[14] Malgrange B., Sur les déformations isomonodromiques. I. Singularités régulières, in Mathematics and Physics
(Paris, 1979/1982), Progr. Math., Vol. 37, Birkhäuser Boston, Boston, MA, 1983, 401–426.
[15] Malgrange B., Déformations isomonodromiques, forme de Liouville, fonction τ , Ann. Inst. Fourier (Greno-
ble) 54 (2004), 1371–1392.
https://doi.org/10.1007/BF01192828
https://doi.org/10.1007/BF01192828
https://arxiv.org/abs/math.FA/9909010
https://doi.org/10.1007/s00220-009-0961-7
https://doi.org/10.1007/s00220-009-0961-7
https://arxiv.org/abs/0902.4716
https://arxiv.org/abs/1601.04790
https://doi.org/10.1093/imrn/rnr066
https://arxiv.org/abs/1005.4083
https://doi.org/10.1007/BF01192827
https://doi.org/10.1007/BF01192827
https://arxiv.org/abs/math.CA/9907165
https://doi.org/10.1007/978-3-0348-5492-4
https://doi.org/10.1007/978-3-0348-5492-4
https://arxiv.org/abs/1608.00958
https://doi.org/10.1007/s002200200614
https://doi.org/10.1007/s002200200614
https://arxiv.org/abs/solv-int/9706002
https://arxiv.org/abs/1604.03082
https://doi.org/10.1093/imrn/rnu209
https://arxiv.org/abs/1403.1235
https://doi.org/10.1142/S0217979290000504
https://doi.org/10.5802/aif.2052
https://doi.org/10.5802/aif.2052
1 Introduction
2 Construction of the Fredholm determinants
3 The SL2 case
3.1 Fredholm determinants for different factorizations
3.1.1 Determinants for different choices of
3.1.2 Determinants for different factorizations
4 Conclusion
References
|