Embed Size (px)
Citation preview
Ding-Iohara-Miki symmetry of network matrix models
A. Mironova,b,c,d,∗, A. Morozovb,c,d,†, and Y. Zenkevichb,d,e,‡
FIAN/TD-07/16IITP/TH-05/16
ITEP/TH-06/16INR-TH-2016-008
a Lebedev Physics Institute, Moscow 119991, Russiab ITEP, Moscow 117218, Russia
c Institute for Information Transmission Problems, Moscow 127994, Russiad National Research Nuclear University MEPhI, Moscow 115409, Russia
e Institute of Nuclear Research, Moscow 117312, Russia
Abstract
Ward identities in the most general “network matrix model” from [1] can be described in terms ofthe Ding-Iohara-Miki algebras (DIM). This confirms an expectation that such algebras and their variouslimits/reductions are the relevant substitutes/deformations of the Virasoro/W- algebra for (q, t) and (q1, q2, q3)deformed network matrix models. Exhaustive for these purposes should be the Pagoda triple-affine ellipticDIM, which corresponds to networks associated with 6d gauge theories with adjoint matter (double ellipticsystems). We provide some details on elliptic qq-characters.
1 Introduction
Recently, basing on the previous studies in [2]-[11], we introduced [1] a generic Dotsenko-Fateev (DF) [4] networkconformal matrix model, associated with the most general brane web/network (the low-energy limit of toricCalabi-Yau compactifications). The first question to ask about this theory is what is the set of the relevant“Virasoro/W- constraints”: the Ward identities, which are satisfied by its partition function. In this paper, weargue that the substitute/deformation of the CFT stress tensor, which generates these identities, is now providedby the analogues of the q-characters [12] in the elliptic Ding-Iohara-Miki algebra (DIM) [13, 14], as anticipatedat different deformation levels in [15]-[19].
We remind [2] that there are three equivalent ways to derive Ward identities in matrix models (and otherquantum field and string theory models):
(i) by making a change of integration variable [20],(ii) by considering an average of a total derivative [21] and(iii) by building a matrix model from a free field correlator with a given symmetry [3, 22].
The first two methods can seem identical, but in fact this is not quite true: (ii) is technically simpler (morestraightforward) than (i), but instead the emerging algebraic structure is more difficult to reveal. Ideal for thistask is the method (iii), which we now briefly remind.
Given a symmetry generating operator (or a set of operators) T (say, the stress tensor and higher W -algebragenerators), one gets a set of identities⟨
Ψ∣∣∣ G(p) T Q
∣∣∣vac⟩
= 0 ⇐⇒ L(∂p)⟨
Ψ∣∣∣ G(p) Q
∣∣∣vac⟩
= 0 (1)
where• |vac〉 is a “vacuum” state annihilated by T , T |vac〉 = 0,• Q is any “screening” operator which commutes with T , [T , Q] = 0,
∗[email protected]; [email protected]†[email protected]‡[email protected]
1
arX
iv:1
603.
0546
7v4
[he
p-th
] 2
2 M
ar 2
016
• 〈Ψ| is an arbitrary state usually made out of vertex operators, and• G(p) is an intertwiner with the property L(∂p)G(p) = G(p)T , which can be used to convert operator(s) T
into differential/difference operators L acting on “the time variables” p.This is a very general group theoretical construction describing a partition function
Z(p) =⟨
Ψ∣∣∣ G(p) Q
∣∣∣vac⟩
(2)
with a given (T -induced) symmetry as a matrix element. Conformal matrix models [3] arise in this way whenmatrix elements are correlators of 2d free fields and integrals or sums over partitions (interpreted as matrixmodel eigenvalue integrals/sums) emerge from an explicit description of screening charges Q (which are thecentralizer of T ) in the free field Fock space.
mf,2 −mf,1
a(1)
a(2)
mf,2 −mf,1
mf,2+mf,1
2
mf,2+mf,1
2
mbif
Λ1
Λ2
NS5 NS5 NS5
D4
D4
a)
Qm,1
Qm,2
QF,1
QB,1
QF,2 QB,2
Qm,4
Qm,3
QF,3
b)
SU(2) SU(2)2× 2 2× 2
c)
Figure 1: a) Type IIA brane diagram consisting of two horizontal and three vertical intersecting lines representing NS5 and D4branes. The low energy theory in this background is 4d N = 2 gauge theory with SU(2)2 gauge group. Λi are exponentiatedcomplexified gauge couplings, a(a) are Coulomb moduli and ma are the hypermultiplet masses. b) The toric diagram of theCalabi-Yau threefold, corresponding to the 5d gauge theory with the same matter content. Edges represent two-cycles withcomplexified Kahler parameters Qi, which play the same role as the distances between the branes in a). c) The quiver encoding thematter content of the gauge theory. SU(2) gauge groups live on each node and bifundamental matter on each edge. The squaresrepresent pairs of (anti)fundamental matter hypermultiplets.
Reversing the logic, one can start from generic network matrix model [1], associated with the toric diagramin Fig.1 b),
ZΓ(q, t|p) =∑{RE}
∏V
CR(E′V ),R(E′′V ),R(E′′′V )(q, t|pV ) (3)
associated with a planar 3-valent graph Γ with triples of edges E′V , E′′V , E
′′′V merging at vertices V . The sum goes
over Young diagrams RE on the edges and the topological vertices C are provided [23, 24, 8] by weighted sums over3d partitions with given boundary conditions R(E′V ), R(E′′V ), R(E′′′V ). The weights depend on compactificationparameters q1,2,3 and on auxiliary time variables pV (their background values can be used to develop thecheck-operator formalism a la [26]). Usually, time variables are ascribed to edges, not vertices (and we do so
2
in (4) below, but the right procedure remains disputable – and in (3) we absorb propagators into vertices tosimplify the formula. Two of the deformation parameters are also parameterized as q = eε1 , t = e−ε2 and thethird one is put equal to q3 = t/q in 5d theory, while in 6d theories it remains free. In fact, at least, at thealgebraical level, more Kerov parameters [25] of the same nature can also be included [24]. Especially importantis the elliptic case, where infinite set of Kerov parameters form a geometric progression. For generic deformationthe model belongs to the universality class [27, 28] of the double elliptic integrable systems [29] and is invariantunder a rich set of large canonical transformations (dualities). As to the infinitesimal transformations, they alsoform an interesting closed algebra of Ward identities, which can be revealed by the following sequence of stepsbriefly encountered above:
• Breaking horizontal/vertical symmetry. Rewrite ZΓ as an eigenvalue matrix integral/sum of the
form (∏′
means that the multiplier ∆(x
(a)i,α, ~x
(a)i,α
)is excluded from the product)
ZΓ`(q|p) =
∫ [∏α,a
d~x(a)α
]∏′α,βa,b
∆(~x(a)α , ~x
(b)β
)Cabαβexp
∑α,a,k
pak,αuk(~x(a)α )
(4)
Here (Jackson like) integrals over ~xα = {x(a)α,i, i = 1, . . . , N
(a)α , α = 1, . . . ,m(a), a = 1, . . . , n} substitute
the sums over Young diagrams for “vertical” edges E(a)vert of the web-diagram Γ (the choice is actually
dictated by the 5d/4d limit and N(a)α may be interpreted as the number of lines in the Young diagram, which
can be arbitrary). In the example from Fig. 1 we have n = 2, m1 = m2 = 2. We put an additional index onΓ to remind about additional vertical/horizontal (quiver) structure on the graph Γ, implicit in the formula.The set of “Casimir” functions uk(~x) is usually adjusted to simplify the differential/difference equations (7)below (we will actually use the Miwa transform, converting Casimirs into vertex operator insertions). The
sums over diagrams for “horizontal” edges E(ab)hor are substituted by the q1,2,3-dependent Vandermonde-like
quantities ∆(~x(a)α , ~x
(b)β ) which can be realized as a free field pairwise correlator of screening currents
S(a)α (x), and the product arises as a consequence of the Wick theorem.
• Screening operators. Using the screening charges given by single-variable integrals (perhaps, Jacksonsums) of the screening currents (which are exponentials of free fields)
S(a)α =
∮Ca
dx : exp
(∑k∈Z
xkak,α
): [an,α, am,β ] = ξnδn+m,0δαβ (5)
one can realize the matrix model integral as an average
ZΓ (p) =
⟨0
∣∣∣∣∣ G(p)
Q︷ ︸︸ ︷∏α,a
(S(a)α
)N(a)α
∣∣∣∣∣0⟩N
(6)
where the vacuum state is N =∑αN
(a)α -charged vacuum with respect to the Heisenberg operators an,α and
G(t). The number of free fields {an,α} actually depends on the number of horizontal edges (“D-branes”)in the original graph Γ, i.e. on the ranks of gauge groups in 4d version of the model (4). We implicitlyinclude N into the set of time-variables.
• Ward identities satisfied by the matrix integral can be described in two ways. In the free field terms theyare provided by the free field operators Wa that are defined to commute with the screening charges, whilein terms of time variables (i.e. literally as a set of constraints imposed on the time-dependent integral)they are expressed with the help of the intertwiner G(p):
Wa(p, ∂p)ZΓ (p) =⟨
0∣∣∣ G(p) Wa Q
∣∣∣0⟩N
= 0 (7)
The question is what is the algebra formed by this set of constraints on a matrix integral. In simplestexamples this is just a Borel subalgebra of Virasoro or various Wma algebras, where ma are related by thenumber of horizontal edges in Γ (e.g. m1 = m2 = 2 in Fig. 1).
3
• Toroidal algebra. One can embed all the Wma-algebras associated with the set of matrix integralsof a given type into a larger algebra. For instance, in the case of Dotsenko-Fateev integrals associatedwith Nekrasov functions and topological vertices corresponding to all graphs, this gives rise to toroidalalgebras: affine Yangians in the case of 4d Nekrasov functions [30, 19], Ding-Iohara-Miki (DIM) (quantumtoroidal) algebra in the 5d case [13, 14, 31] and elliptic DIM algebra in the 6d case [32, 9]. Concrete quivercorresponds to a set of representations of the toroidal algebra given by a fixed number of Young diagrams.Moreover, the (refined) topological vertex can be obtained as a matrix element of the intertwining operatorsof the DIM algebra [33]. We concentrate below on the level 1 representation of DIM algebra so that therealways exists a simple bosonization [32]. For generic levels an analogue of the free-field representation ofKac-Moody algebras [34] will be needed.
• qq-characters. Generalized stress tensor operators Wa can be actually understood in terms of the DIM R-matrices, and from this perspective they give abstract algebraic description of the qq-characters [16, 17, 18].This construction generalizes ordinary q-characters for quantum groups introduced in [12].
• Systems of symmetric functions. One can associate with the set of matrix integrals/algebra a set ofsymmetric functions in two different ways. One option is to construct them directly from the integral,omitting one set of integrations [35, 36]. In the simplest example of the matrix integral with n = 1 we fixthe Young diagram λ with lengths of lines λα, α = 1 . . . (m− 1) and the corresponding symmetric function
of variables xi ≡ x(N)i corresponding to λ is given by the matrix integral
PλT (xi) ∼∫ [m−1∏
α=1
λα∏i
dx(α)i
]m∏
α,β=1
∏′λα,λβ
i,j∆(~xα, ~xβ
)Cαβ(8)
where λm is put equal to m. This is a generalization of old formulas from [37] and and it can be consideredas an extension of the degenerate field insertion into the conformal block [38] in the DF approach.
Another way to construct symmetric functions [36, 10] is to consider a level 1 representation of the algebraso that it is realized by the Heisenberg algebra. Choose a Hamiltonian as an element of the algebra, itis a function of generators an. Realizing them in terms of time variables, an<0 ∼ pn, an>0 ∼ ∂pn , oneobtains a set of eigenfunctions of the Hamiltonian as functions of pn. After the Miwa transformation,pn =
∑i x
ni they give rise to symmetric functions of xi. For instance, the level one representation of the
DIM gl1 algebra leads to the set of gl1 Macdonald polynomials. As a next step, one can consider the setsof eigenfunctions which diagonalize co-products (of degree N − 1) of the Hamiltonian (i.e. representationsof higher levels), which, in this concrete example, leads to the generalized glN Macdonald polynomials.
• Lift to the graph level. The next step is restoration of the vertical/horizontal symmetry and liftingthe symmetry (Ward identities) to the original network matrix model (4). Important at this step is thattopological vertices are associated with matrix elements of the intertwining operators of the DIM algebra[33].
The crucial ingredient of this construction is the centralizer of the algebra of constraints, defining the screeningoperators and the matrix model. The centralizer depends on the representation, and it is this dependence thatleads to a variety of different matrix models, encoded by the graph Γ. Once the graph (with some additionaldecorations: preferred direction, etc.) is chosen, the particular representation of DIM is fixed and so is theparticular matrix model. However, traces of the larger DIM symmetry remain in various forms, the most notableexample being the spectral duality [39, 11, 40] connecting multi-matrix models with different numbers of matricesand vertex operator insertions.
In the simplest case, associated with 4d Seiberg-Witten theory, the role of T is played by the stress tensorT (z) = 1
2∂φ(z)2 +Q∂2φ(z), which generates the ordinary Virasoro algebra (and its WN -algebra generalizations),and Z(t) is just the ordinary Dotsenko-Fateev (DF) matrix model of [4]. Various types of q/t/q123-deformations,associated with reviving of the hidden compactification moduli, i.e. revealing the hidden 6d and M-theory natureof the theory, require a lifting/resolution of 1
2∂φ(z)2 +Q∂2φ(z) of a peculiar Toda-like combination of vertexoperators:
T (z) = : eΦ(z)e−Φ(t−1z) : + t : e−Φ(tz/q)eΦ(z/q) : (9)
where
Φ(z) =∑n≥1
zn
nα−n + Φ0 −
∑n≥1
z−n
nαn, (10)
4
and the modes of Φ(z) satisfy the q-deformed commutation relations:
[αn, αm] =n
1 +(qt
)|n| 1− q|n|1− t|n|δn+m,0. (11)
Given these q-boson relations, T (z) generates the q-deformed Virasoro algebra Virq,t. Deformed stress-energytensor (9) can be guessed from the requirement that it commutes with the screening current. The expression forthe screening current essentially determines the matrix model and its symmetry. Concretely, the q-deformedscreening current is given by
S(x) = :∏k≥0
exp
(−Φ(qkx) + Φ(qktx) + Φ(qk+1x)− Φ
(qk+1
tx
)): (12)
We will derive the formulas for T and S in detail in sec. 2.2.5, in particular we obtain the form (9) of stress-energytensor in Eq. (48).
One can see that the expressions for T and S are not symmetric under the exchange of t and q−1, which is,as we will see, the natural symmetry of the DIM algebra. This is another artifact of the choice of a concreterepresentation/matrix model description of the object with larger symmetry. All the essential quantities ofeach particular model should be symmetric w.r.t. q ↔ t−1, though the intermediate results do not respect thissymmetry.
In the double-scaling limit q = e~ → 1, t = qβ the ordinary Virasoro stress-energy tensor T (z) = 12∂φ(z)2 +
Q∂2φ(z) (with Q =√β −√β−1
) is recovered from (9):
T (z)→ 2 + ~(1− β) +~2
2
[(β − 1)2 + z2T (z)
]+O(~3), (13)
where
φ(z) = 2√β limq,t→1
t=qβ
Φ(z) =∑n≥1
zn
nα−n + Φ0 −
∑n≥1
z−n
nαn, (14)
and αn are ordinary boson generators, satisfying [αn, αm] = 2nδn+m,0.As already mentioned, a nice bonus is that multi-field generalization of (9), which in 4d leads to substitution
of Virasoro by W -algebras, is now just another representation of the same symmetry algebra. In other words,after the deformation the Sugawara-like bi- and multi-linear combinations of currents can be obtained fromcomultiplication of the deformed current algebra, without a need to consider its universal enveloping.
The purpose of this paper is a sketchy survey of this remarkable DIM symmetry of (4). Various details willbe presented in separate publications. We will discuss here the 5d and 6d DIM gl1 algebras which correspond tothe quiver gauge theories with fundamental matter. The most interesting case of the DIM affine algebras whichdescribe, in particular, the 6d gauge theory with adjoint matter and correspond to the double elliptic systemswill be touched only briefly. This issue, and also various details of other cases will be presented in separatepublications.
2 A1 (q, t)-matrix model
Let us start with the prototypical example of the A1 (q, t)-deformed conformal matrix model. This is the simplestmodel where q-Virasoro symmetry arises and, therefore, serves as an accessible port of entry to the land of DIMalgebras.
In this section we describe the general scheme for investigating a network-type matrix model. We start bywriting down the conventional definition of the model in terms of matrix integral. However, one should remember,that this is just a particular representation of the network of topological vertices, as in Eq. (3). We next describethe algebraic face of the matrix model more concretely by specifying the screening operators, which OPE givesthe actual matrix model integrals. The centralizer of the screenings inside the representation of DIM gives theW -algebra corresponding to the matrix model, which also generates the qq-characters in the gauge theory. Thisdescription was used in [18] to introduce the W -algebras corresponding to an arbitrary (affine) ADE-type quiver.Our aim in this paper is more general (though in this section we study it on a very humble example). We wouldlike to elucidate the hidden symmetries, which are only visible in the network-type formalism (3) (see [11] for anexample of such an approach). The symmetries of the network/topological string/toric diagram are described
5
by DIM algebra, of which different W -algebras are only particular representations/subalgebras. In this part ofthe paper we will demonstrate explicitly how various concepts in matrix models and gauge theories, such asqq-characters and generalized Macdonald polynomials, are tied together with the help of the DIM algebra.
We introduce the DIM algebra generators and relations in sec. 2.2.1. We describe the simplest representationsof DIM algebra in sec. 2.2.3 and show how they give rise to generalized Macdonald polynomials. In sec. 2.2.5with the help of dressing operators, we build the deformed Virasoro subalgebra of the DIM algebra and show itsconnection to qq-characters in the gauge theory. In sec. 2.2.6 we focus on the details of the dressing procedureand identify it with the reduction of the “U(1) part” in the Nekrasov function/conformal block. We also describethe relation with Benjamin-Ono integrable system.
2.1 Free-field description
The matrix model can be described in two different ways: as a Jackson or contour integrals respectively. Herewe adopt the latter form:
ZA1=
∮dNx∆(q,t)(x)V1(z1, x) · · ·VM (zM , x), (15)
where
∆(q,t)(x) =∏i 6=j
(xixj
; q)∞(
t xixj ; q)∞
,
Va(za, x) =
N∏i=1
(q1−va za
xi; q)∞(
zaxi
; q)∞
, (16)
and the Pochhammer symbol (q-exponential) is defined as (x; q)∞ =∏k≥0(1− qkx). Time variables are traded
for a product of vertex operators V (z): this can be understood/interpreted as a Miwa transform.Following the general recipe given in the introduction, we would like to interpret the matrix model as an
average of screening currents S(x). One can see explicitly that the necessary choice is
S(x) = : exp
−∑n≥1
xn
n
1− tn
1− qn(
1 +(qt
)n)α−n +
∑n≥1
x−n
n
1− t−n
1− q−n
(1 +
(qt
)−n)αn
: (17)
where Φ(x) is defined in Eq. (10). From the q-boson commutation relations (11) one get the following OPE forthe screenings currents
S(x1)S(x2) =
(x2
x1; q)∞
(qtx2
x1; q)∞(
tx2
x1; q)∞
(tx2
x1; q)∞
: S(x1)S(x2) :integer β−→
β−1∏k=0
(1− qk x1
x2
)(1− qk x2
x1
): S(x1)S(x2) :
(18)From the OPE (18) we can immediately see that the matrix model (15) is indeed the correlator of screeningswith vertex operators:
ZA1= 〈0|
∮dNx
N∏i=1
S(xi)V1(z1) · · ·VM (zM )|0〉 (19)
What are the Ward identities for the (q, t)-matrix model? To obtain them let us perform the steps we discussedin the Introduction: first, we introduce time variables pk into the matrix integral inserting into the average (19)the operator
G(p) = exp(∑k>0
pkα−k
)(20)
and, second, we verify that the deformed stress-energy tensor T (z) (9) commutes with the integral of thescreening current (17),
∮S(x)dx/x. This means that T (z) commutes with S(x) up to total derivative (or total
6
q-difference). The OPE of T with S is given by
T (z)S(x) =1− txz1− x
z
: eΦ(z)−Φ(z/t)S(x) : +t1− q
txz
1− q xz: e−Φ(tz/q)+Φ(z/q)S(x) :=
= :
{(1− txz1− x
z
eΦ(z)−Φ(x)+Φ(x/t)−Φ(z/t) + t1− 1
txz
1− xz
eΦ(tx/q)−Φ(tz/q)+Φ(z/q)−Φ(x/q)
)eΦ(x)−Φ(x/t)+
+ (qx∂x − 1)
(t1− 1
txz
1− xz
e[Φ(tx/q)−Φ(tz/q)+Φ(z/q)−Φ(x/q)]+Φ(x)−Φ(x/t)
)}S(x) : (21)
Remarkably, the pole at z = x is exactly canceled in both terms in the first line: the shift in the infinite sumof operators inside S(x) plays a crucial role in this cancellation, and only the total difference remains singular.Since the poles are canceled up to total q-difference, the commutator with screening charge, i.e. with the integralof S(x) vanishes. This fact was used in [18] to derive the regularity of the qq-characters.
This implies that T (z) is a symmetry of the model: negative modes of its Laurent expansion in z annihilatethe vacuum and thus annihilate the entire matrix integral. Inserting the T -S OPE into the matrix integral, weget:
Tpk(z)ZA1(p) =⟨T (z)
⟩=〈0|G(p)V1(z1) · · ·VM (zM )T (z)
∮dNx
∏Ni=1 S(xi)|0〉
〈0|V1(z1) · · ·VM (zM )∮dNx
∏Ni=1 S(xi)|0〉
=
∮dNx∆(q,t)(x)
(∏i
V1(z1, xi) · · ·VM (zM , xi)U(xi, p)
)(N∏i=1
1− txiz1− xi
z
+ P (z|{za}, {va})N∏i=1
1− q xiz1− q
txiz
)= Pol(z)
(22)
where P (z|{za}, {va}) is the contribution of vertex operators, a z-polynomial factor and the time dependence ofthe partition function is encoded in the potential
U(x, p) = exp
−∑n≥1
xn
n
1− tn
1− qn(
1 +(qt
)n)pn
(23)
The deformed stress-energy tensor, written in the bosonized form, as in Eq. (9), or in the form of matrix modelaverage, can also be realized as a difference operator upon identification
α−n = pn, αn =n
1 +(qt
)n 1− qn
1− tn∂
∂pn(24)
leading to a difference equation on the partition function, a counterpart of the Baxter equation. It is sometimecalled a qq-character [16, 18, 17, 19], since it can be considered as a deformation of the Frenkel-Reshetikhinq-character [12] (trace over Cartan part of the quantum R-matrix). Virasoro symmetry of the matrix modelimplies that this average has no negative modes in its z-expansion, i.e. is regular (and therefore polynomial) in z:
regularity of the qq-character = polynomiality of the average 〈T (z)〉 =
= Ward identity (DIM/Virasoro constraint) (25)
While obviously following from commutativity of T (z) with S, this looks like a non-trivial property of the r.h.s.in (22).
Also qq-characters can be thought of as the recurrence relation on the matrix model correlators, obtained byexpanding the average of T (z) in powers of z. The recurrence relations can also be derived by considering thevanishing total difference under the matrix model integral [11]. Of course, this only means that the commutatorof T (z) and S(x) is given by the corresponding total difference. In the case at hand the relevant total difference
7
is given by
0 =
∮dNx
N∑i=1
1
xi
(1− qxi∂i
) xiz − xi
∏j 6=i
xi − txjxi − xj
∆(q,t)(x)
=
=
∮dNx
N∑i=1
1
z − xi
∏j 6=i
xi − txjxi − xj
− tN−1q
z − qxi
∏j 6=i
txi − xjxi − xj
∆(q,t)(x) =
=
∮dNx
N∏j=1
1− txjz
1− xjz
+ t2N−1q
N∏j=1
1− qxjtz
1− qxjz
−QN (z)
(26)
where QN (z) is degree N polynomial in z and xi, and in the last line we have summed over poles in z toobtain the products. The identity (26) is precisely the regularity constraint on the qq-character telling that〈T (z)〉 = 〈P (z)〉 = regular in z. For details of derivation along this route see [11]. We will employ similartechnique to get the symmetry constraints for the elliptic matrix model in sec. 3.
In the next section we show how to obtain the deformed energy-momentum tensor from the representation ofthe abstract DIM algebra.
2.2 Abstract algebraic description
We now describe the algebraic structures of DIM algebra governing the network-type matrix model. Let us
first recall the definition of the DIM algebra Uq(gl1) and its simplest representations and then demonstrate the
connections of this algebra with deformed Virasoro algebra, qq-characters, generalized Macdonald polynomialsand integrable systems.
2.2.1 DIM algebra
This looks like a deformation of the affine quantum algebra Uq(gl2) with the positive/negative root generatorsx±(z), two exponentiated Cartan generators ψ±(z) and the central element γ.
Commutation relations are
G∓(z/w)x±(z)x±(w) = G±(z/w)x±(w)x±(z)
[x+(z), x−(w)] =(1− q)(1− t−1)
1− q/t
(δ(γ−1z/w)ψ+(γ1/2w) − δ(γz/w)ψ−(γ−1/2w)
)ψ±(z)ψ±(w) = ψ±(w)ψ±(z) (27)
ψ+(z)ψ−(w) =g(γw/z)
g(γ−1w/z)ψ−(w)ψ+(z)
ψ+(z)x±(w) = g(γ∓1/2w/z)∓1 x±(w)ψ+(z)
ψ−(z)x±(w) = g(γ∓1/2z/w)±1 x±(w)ψ−(z)
Symz1,z2,z3
z2z−13 [x±(z1), [x±(z2), x±(z3)]] = 0
DIM algebra is a Hopf algebra with comultiplication
∆(ψ±(z)
)= ψ±(γ
±1/22 z) ⊗ ψ±(γ
∓1/21 z)
∆(x+(z)
)= ψ−(γ
1/21 z) ⊗ x+(γ1z) + x+(z) ⊗ 1 (28)
∆(x−(z)
)= 1 ⊗ x−(z) + x−(γ2z) ⊗ ψ+(γ
1/22 z)
where γ±1/21 = γ±1/2 ⊗ 1, γ
±1/22 = 1 ⊗ γ±1/2. The function g(z) = G+(z)
G−(z) is restricted by the requirement
g(z) = g(z−1)−1 and δ(z) =∑n∈Z z
n. We omit expression for the counit and antipode, since we will not needthem.
8
2.2.2 Specification of the structure function
The structure of the algebra is encoded in the function G(z) which is often chosen to be cubic in z with additionalrestriction q1q2q3 = 1:
G±(z) = (1− q1z)(1− q2z)(1− q3z) =(1− q±1z
) (1− t∓1z
) (1− (t/q)±1z
)(29)
Without any harm to commutation relations and comultiplication it can be further promoted to unrestricted q123-and more general Kerov deformations, and even to elliptic function, though details of bosonization procedurebelow should still be worked out in these cases. We describe the elliptic version in sec. 3.
2.2.3 Level one Fock representation
The simplest representation of DIM algebra is the level one representation ρu acting on the Fock moduleFu, generated by the q-deformed Heisenberg creation operators a−n from the vacuum |u〉 annihilated by theannihilation operators an. The Heisenberg generators satisfy
[an, am] = n1− q|n|
1− t|n|δn+m,0 (30)
Note that an are normalized differently from αn in eqs. (10), (11) (that normalization was chosen to maximallysimplify the final expressions). Of course the an generators are related to αn generators in a simple way:
αn =1
1 + (q/t)nan n ≥ 1 (31)
α−n = a−n n ≥ 1 (32)
The generators of the DIM algebra are expressed in terms of the Heisenberg generators:
ρu(x+(z)
)= uη(z) = u : exp
∑n≥1
1− t−n
na−nz
n −∑n≥1
1− tn
nanz−n
:
ρu(x−(z)
)= u−1ξ(z) = u−1 : exp
∑n≥1
1− t−n
n
(t
q
)n/2a−nz
n −∑n≥1
1− tn
n
(t
q
)n/2anz−n
:
ρu(ψ±(z)
)= ϕ±(z) = exp
∓∑n≥1
1− t±n
n
(1−
(t
q
)n)a±nz
∓n
(33)
ρu(γ) =
(t
q
)1/2
Let us see an example how OPE of these operators reproduces the DIM commutation relation:
η(z)η(y) =
(1− z
y
)(1− q
tzy
)(
1− 1tzy
)(1− q zy
) : η(z)η(y) :=
(1− q
tzy
) (1− 1
tyz
) (1− q yz
)(1− 1
tzy
)(1− q zy
) (1− q
tyz
)η(z)η(y) =G−
(zy
)G+
(zy
)η(z)η(y) (34)
2.2.4 Level two Fock representation and generalized Macdonald polynomials
Tensor product of m Fock representations Fu1 ⊗ · · · ⊗ Fum can be easily obtained from the comultiplication (28)and will be called the level m Fock representation. In this tensor product the generators of DIM algebra are
expressed in terms of m q-Heisenberg generators a(a)n , a = 1, . . . ,m. In particular, we will need the expression
for x+(z) in this representation:
ρ(2)u1,u2
(x+(z)) = u1Λ1(z) + u2Λ2(z) = u1η1(z) + u2ϕ−1
((t/q)
1/4z)η2
((t/q)
1/2z). (35)
where we use the shorthand notation Λ1,2 for the components of the level two representation ρ(2)u1,u2 = (ρu1⊗ρu2)∆,
and the subscript denotes the number of term in the tensor product, e.g. η1(z) = η(z)⊗ 1.
9
There is an distinguished basis in Fu1⊗ · · · ⊗ Fum , the basis of generalized Macdonald polynomials [10]
obtained by diagonalizing the action of the zero mode of x+(z). Representation of this zero mode was calledgeneralized Macdonald Hamiltonian in:
Hgen1 = ρ(2)
u1,u2(x+
0 ) =
∮C0
dz
zρ(2)u1,u2
(x+(z)), (36)
In those papers the following definition of the generalized Macdonald polynomials was given:
Hgen1 MAB(a
(1)−n, a
(2)−n)|u1 ⊗ u2〉 = [u1κA(q, t) + u2κB(q, t)]MAB(a
(1)−n, a
(2)−n)|u1 ⊗ u2〉, (37)
whereκA = (1− t)
∑i≥1
qAit−i. (38)
These polynomials were instrumental in demonstrating the 5d version of the AGT conjecture [41]. Matrixelements of Virasoro primary fields in this basis turned out to coincide with fixed point contributions in theNekrasov partition function. Thus, after decomposition of conformal blocks in terms of generalized Macdonaldpolynomials, the AGT relation becomes explicit. In the 4d limit this special basis degenerates into the basis ofgeneralized Jack polynomials [10], with similar properties.
2.2.5 W -algebra, Ward identities and qq-characters from DIM
As we have announced in the introduction, the great benefit of DIM approach is that it describes different matrixmodels from a unified viewpoint. In particular, m-multimatrix models have Wm-algebra symmetries, and thesealgebras are all particular representations of subalgebras of DIM algebra.
q-deformed Wm-algebra, which is also called Wq,t(slm), is obtained from level m Fock representation of theDIM algebra as follows. The stress-energy tensor of the Wm-algebra is obtained from the dressing of the x+
generator of DIM. More concretely, we have:
t(z) = A(z)x+(z)B(z), (39)
where
A(z) = exp
−∑n≥1
1
γn − γ−nb−nz
n
, B(z) = exp
∑n≥1
1
γn − γ−nbnz−n
(40)
and bn are the modes of the ψ± generators:
ψ±(z) = ψ±0 exp
±∑n≥1
b±nγn/2z∓n
. (41)
The stress-energy tensor T of the WM -algebra is the representation of the dressed current t(z) in the level mFock module. For the Virasoro case (m = 2), using Eq. (35), we get
T (z) = ρ(2)u1,u2
(t(z)) = u1Λ1(z) + u2Λ2(z) =
= ρ(2)u1,u2
(A(z))(u1η1(z) + u2ϕ
−1
((t/q)
1/4z)η2
((t/q)
1/2z))
ρ(2)u1,u2
(B(z)). (42)
where Λi(z) are dressed versions of the components Li(z). From Eq. (42) we see that T (z) depends on twosets of Heisenberg generators (hidden inside η1, η2 and ϕ−) acting on the tensor product of two Fock modules.However, as we will see explicitly in the next section, the expression for T actually depends only on one linear
combination of a(1)n and a
(2)n . Related to this fact is that in the level two representation the product of Λ1,2
elements is equal to identity:: Λ1(z)Λ2 (zq/t) : = 1. (43)
To see this fact we should write explicit (though lengthy) expressions for Λ1,2 in the level two representation:
Λ1(z) = : exp
∑n≥1
1
n
1− t−n
1 + (q/t)n z
n(α
(1)−n − (q/t)
n/2α
(2)−n
)−∑n≥1
1− tn
nz−n
(α(1)n − (q/t)
n/2α(2)n
) : (44)
Λ2(z) = : exp
−∑n≥1
1
n
1− t−n
1 + (q/t)n (zt/q)
n(α
(1)−n − (q/t)
n/2α
(2)−n
)+∑n≥1
1− tn
n
(z−1q/t
)n (α(1)n − (q/t)
n/2α(2)n
) :
(45)
10
From these expressions we see that indeed : Λ1(z)Λ2 (zq/t) : = 1. We also identify the combinations of creationand annihilation operators, on which T depends, and denote these combinations by α. They are given by
α−n =1
1 + (q/t)n
(α
(1)−n − (q/t)
n/2α
(2)−n
), n ≥ 1 (46)
αn =(α(1)n − (q/t)
n/2α(2)n
), n ≥ 1 (47)
One can see that the commutation relations for αn are the same as for α(1)n . Now Λ1,2 and the stress-energy
tensor T are all nicely written in terms of these combinations:
T (z) = u1 : eΦ(z)e−Φ(t−1z) : +u2 : e−Φ(tz/q)eΦ(z/q) : (48)
where the definition of Φ is similar to that of Φ from Eq. (10), only the role of bosons αn is now played by αn.The constants u1 and u2 can be absorbed into the definition of zero modes, which brings Eq. (48) into the formof the deformed stress-energy tensor identity (9). The zero modes of the screening operators are omitted tosimplify the formulas.
Finally, the operators αn are in fact precisely those bosonic operators, in terms of which we have defined ourmatrix model (19). We have, therefore, identified the Ward identities/Virasoro constraints of the matrix modelwith the particular combination of the DIM operators in the level two Fock representation. We can also makethe identification with qq-character more explicit by introducing the usual notation:
Λ1(z) = Y(z), Λ2(z) = Y−1
(t
qz
). (49)
The last definition follows from the condition (43). Now we would like to understand where the other combinationof the bosonic generators is hidden. To see this we have to revisit the dressing procedure, for the current t(z).
2.2.6 Virq,t ⊕ Heisq,t reduction is equivalent to dressing
In this section we show how the dressing operators α(z) and β(z) are in fact performing the reduction of thealgebra Virq,t ⊕ Heisq,t acting in the level two Fock representation to its Virq,t part. The condition (43) canbe thought of as a gauge condition used to kill the Heisq,t degrees of freedom, which enter both Λ1 and Λ2
multiplicatively. This separation of variables is usual for description of Hamiltonian reductions in the free fieldformalism [42].
To this end let us look at the bosonization of the dressing operators A(z) and B(z). From Eqs. (33), (40) weget
ρ(2)u1,u2
(A(z)) = exp
−∑n≥1
1
n
1− t−n
1 + (q/t)n (q/t)
n/2zn(
(q/t)n/2
α(1)−n + α
(2)−n
) , (50)
ρ(2)u1,u2
(B(z)) = exp
∑n≥1
1− tn
n(q/t)
n/2z−n
((q/t)
n/2α(1)n + α(2)
n
) .
In the exponent, these two operators contain precisely the linear combination of α(1,2)n orthogonal to αn. We
denote the new bosons by αn:
α−n =(q/t)
n/2
1 + (q/t)n
((q/t)
n/2α
(1)−n + α
(2)−n
), (51)
αn = (q/t)n/2(
(q/t)n/2
α(1)n + α(2)
n
).
These bosons commute with αn and satisfy slightly modified (compared to (11)) commutation relations amongthemselves:
[αn, αm] =n(1− qn)(q/t)n
(1 + (q/t)n) (1− tn)δn+m,0. (52)
Since the Virq,t algebra is entirely built out of αn, the new generators αn commute with Virq,t and form theadditional q-deformed Heisenberg algebra Heis. One can recall that such a situation is common in the studyof AGT relations [43], where Nekrasov function [44, 45] usually corresponds to the conformal block [46] of the
11
Virasoro algebra times an additional “U(1) factor”, which corresponds to an extra boson, forming the Heisalgebra [47]. Here we get the extra boson for similar reasons: we are working in the tensor product of two Fockmodules, and have to eliminate the “diagonal part” of the bosonized algebra. This elimination corresponds tothe dressing transformation, which is nothing but the transformation to the “center of mass frame” for the two
bosons α(1,2)n .
Finally, we can write down a compact expression for the undressed current x+(z) in the level two Fockrepresentation:
ρ(2)u1,u2
(x+(z)
)= u1Λ1(z) + u2Λ2(z) = T (z)Z(z) = T (z) : eΦ(z)−Φ(z/t) : (53)
where Φ(z) is again the bosonic field defined analogously to (10) using αn generators. We introduced the Heisq,tqq-character Z(z), in terms of which the undressed current factorizes into the product of two terms correspondingto algebras in Virq,t ⊕ Heisq,t.
The factorized form of the current x+(z) is also reflected in structure of its zero mode: the Hgen1 operator.
Written in this form it gives the trigonometric generalization of the Benjamin-Ono (BO) equation [48], thecontinuous integrable model also related to the AGT correspondence. It is easy to see the structure of the BOHamiltonians in the double scaling limit q → 1, t = qβ :∮
C0
dz
zρ(2)u1,u2
(x+(z)) = 2 + ~(1− β) +~2
2(I1 + C1) +
~3β
2(I2 + C2) +O(~4), (54)
where C1,2 are constants,
I1 = L0 + 2∑n≥1
ˆα−n ˆαn −1− 3Q2
6, (55)
I2 =∑k 6=0
ˆα−kLk + 2Q∑n≥1
n ˆα−n ˆαn +1
3
∑n+m+k=0
ˆαn ˆαm ˆαk, (56)
and ˆαn are the ordinary Heisenberg generators, obtained from αn in the double scaling limit. All higher BOHamiltonians appear in the higher terms. What we have found is that generalized Macdonald polynomials are infact joint polynomial eigenfunctions of the quantum BO system.
3 Elliptic DIM algebra and elliptic matrix model
In this section we describe the elliptic generalization of the matrix model and DIM algebra governing it. As wewill see, most of the discussion is exactly parallel to the trigonometric case. This is another manifestation of theuniversality of network type matrix models and the DIM algebra. The fact that the description of the ellipticcase is so similar to the trigonometric one gives one the hope that the corresponding structure in the doubleelliptic case might also be tractable.
3.1 Elliptic matrix model
This matrix model has been described in [9, 1], and we follow the notations of this paper.
ZellA1
=
∮dNx∆
(q,q′,t)ell (x)V1(z1, x) · · ·VM (zM , x), (57)
where
∆(q,q′,t)ell (x) =
∏i 6=j
(xixj
; q, q′)∞
(qq′
txjxi
; q, q′)∞(
t xixj ; q, q′)∞
(qq′
xjxi
; q, q′)∞
, (58)
Va(za, x) =
N∏i=1
(q1−v z
xi; q, q′
) (qq′ xiz ; q, q′
)(zxi
; q, q′) (q′qv xiz ; q, q′
) , (59)
where the double q-Pochhammer symbol is (z; q, q′)∞ =∏k,l≥0(1− zqkq′l).
12
This elliptic integral arises from the following screening currents:
S(x) = :∏k≥0
exp
(−Φ(qkx) + Φ(qktx) + Φ(qk+1x)− Φ
(qk+1
tx
)):=
= : exp
−∑n 6=0
xn
n
1− tn
(1− qn)(1− q′|n|)
(1 +
(qt
)n)α−n +
∑n 6=0
x−n
n
1− tn
(1− qn)(1− q′|n|)
(1 +
(qt
)n)β−n
: (60)
where the bosons αn and βn obey the commutation relations:
[αn, αm] =n(1− q′|n|)1 +
(qt
)|n| 1− q|n|
1− t|n|δn+m,0,
[βn, βm] =nq′|n|(1− q′|n|)
1 +(qt
)|n| 1− q|n|
1− t|n|δn+m,0, (61)
[αn, βm] = 0. (62)
and Φ(z) is the field built out of αn and βn:
Φ(z) =∑n6=0
zn
n(1− q′|n|)α−n −
∑n 6=0
z−n
n(1− q′|n|)β−n (63)
Notice the presence of two sets of boson generators αn and βn, which is related to the modular invariance of theelliptic model. More concretely two bosons produce two terms in the product representation the theta-function:∏k≥0(1− q′kz)(1− q′kq′/z), the elliptic version of the free field correlator (1− z). This explains why the powers
of z in front of αn and βn are opposite, and also why their commutation relation differ by q′|n|.Of course, the stress-energy tensor, which generates the centralizer of the screening charge Q =
∮S(x)dx/x
also depends on two sets of bosonic variables. It is very analogous to the trigonometric case:
T (z) = : eΦ(z)e−Φ(t−1z) : + t : e−Φ(tz/q)eΦ(z/q) : (64)
This elliptic stress-energy tensor generates the elliptic deformation of the Virasoro algebra, which has beenconsidered in many works [9]. We proceed along the lines of the previous section and move to the correspondingDIM algebra, which gives tensor T in the level two representation.
3.2 Elliptic DIM algebra, elliptic Virasoro and ILW equation
Elliptic version of DIM algebra is generated by the same set of operators as the ordinary DIM: x±(z), ψ±(z) andthe central element γ. The relations are a copy of Eq. (27), except for the [x+, x−] relation, which changes to
[x+(z), x−(w)] =Θq′(q; q
′)Θq′(t−1; q′)
(q′; q′)3∞Θq′(q/t; q′)
(δ(γ−1z/w)ψ+(γ1/2w) − δ(γz/w)ψ−(γ−1/2w)
)(65)
where Θp(z) = (p; p)∞(z; p)∞(p/z; p)∞ is the theta-function. Also, most importantly, the structure functionG±(z) is now not trigonometric, but elliptic:
G±ell(z) = Θp(q±1z)Θp(t
∓1z)Θp(q∓1t±1z), (66)
The comultiplication ∆ is exactly the same as in the trigonometric case, given by Eqs. (28). As with the matrixmodel in the previous section, the essential difference with the trigonometric case appears when one tries to buildFock representation of elliptic DIM: one set of bosons turns out not to be enough. We need at least two sets ofHeisenberg generators an and bn to reproduce the commutation relations of the elliptic algebra. Concretely, we
13
have for the level one representation:
ρu(x+(z)) = uη(z) = u : exp
−∑n6=0
(1− tn)z−n
n(1− q′|n|)an
exp
−∑n 6=0
(1− t−n)q′|n|zn
n(1− q′|n|)bn
:
ρu(x−(z)) = u−1ξ(z) = u−1 : exp
∑n 6=0
(1− tn)p−|n|/2z−n
n(1− q′|n|)an
exp
∑n 6=0
(1− t−n)p|n|/2q′|n|zn
n(1− q′|n|)bn
:
ρu(ψ+(z)) = ϕ+(z) = exp
(∑n>0
(1− tn)(p−n/2 − pn/2)p−n/4
n(1− q′n)
(z−nan − p
n2 q′nznbn
))(67)
ρu(ψ−(z)) = ϕ−(z) = exp
(−∑n>0
(1− t−n)(p−n/2 − pn/2)p−n/4
n(1− q′n)
(zna−n − p
n2 q′nz−nb−n
))ρu(γ) = (t/q)
1/2,
where p = qt and the bosons an and bn satisfy the following commutation relations:
[am, an] = m(1− q′|m|)(1− q|m|)
1− t|m|δm+n,0,
[bm, bn] = m(1− q′|m|)(1− q|m|)
(pq′)|m|(1− t|m|)δm+n,0, (68)
[am, bn] = 0.
Again, the fields an, bn are related to αn and βn by a simple redefinition.The dressed current t(z) = A(z)x+(z)B(z), corresponding to the stress energy tensor is given by exactly
the same expression (39), as in the ordinary DIM case. Moreover, the dressing operators A(z) and B(z) areconstructed from the ψ± generators of the elliptic DIM algebra using the same formulas (40) as give above. In
the level two representation ρ(2)u1,u2 the element t(z) produces the elliptic Virasoro stress-energy tensor (64).
Let us also mention that the undressed elliptic DIM charge∮x+(z)dz/z also leads to several very interesting
objects. In the level one representation it gives elliptic Ruijsenaars Hamiltonian, while in the second levelrepresentation it is the difference version of the intermediate long-wave (ILW) Hamiltonian [49], which itself is ageneralization of the Benjamin-Ono system.
3.3 Ward identities and qq-characters
One can derive Ward identities in the same algebraic fashion as for the trigonometric case. The OPE of thestress-energy tensor (64) with the screening current (60) is given by:
T (z)S(x) =Θq′
(txz
)Θq′
(xz
) : eΦ(z)e−Φ(z/t)S(x) : +tΘq′
(qxtz
)Θq′
(qxz
) : e−Φ(tz/q)eΦ(z/q)S(x) : (69)
which is non-singular up to total q-difference due to the same cancellation, as in Eq. (21), and we use the sametime insertion operator (20), but this time depending on two sets of times, pk and pk related to sets of Heisenbergoperators α−n and βn:
G(p) = exp(∑k>0
pkα−k +∑k>0
pkβk
)(70)
Thus the insertion of T (x) into the correlator corresponds to the insertion of the following expression under thematrix model integral:
Tpk(z)ZellA1(p) = 〈T (z)〉 =
∮dNx∆
(q,q′,t)ell (x)
(M∏a=1
N∏i=1
Va(za, xi)U(xi, p, p)
)×
×
N∏j=1
Θq′
(txjz
)Θq′
(xjz
) + P (z|{za}, {va})N∏j=1
Θq′( qxjtz
)Θq′
( qxjz
) = QN (z), (71)
14
where P (z|{za}, {va}) and QN (z) are products of theta functions of the form∏a Θq′(z/λa), and the potential
now has the form
U(x, p, p) = exp
[−∑n>0
xn
n
1− tn
(1− qn)(1− q′n)
(1 +
(qt
)n)pn −
∑n>0
xn
n
1− t−n
(1− q−n)(1− q′n)
(1 +
(t
q
)n)pn
](72)
while the difference realization of the operator Tpk(z) is given by the substitution
α−n = pn, αn =n(1− q′n)
1 +(qt
)n 1− qn
1− tn∂
∂pn, βn = pn, β−n = −nq
′n(1− q′n)
1 +(qt
)n 1− qn
1− tn∂
∂pn(73)
This gives the elliptic qq-character corresponding to the 6d gauge theory corresponding to the A1 quiver, i.e. thegauge group should consist of single SU(n) factor possibly with some fundamental matter hypermultiplets.
As we have seen in the trigonometric case, there is another very explicit way to derive the Ward identities:to consider the vanishing integral of a cleverly chosen total difference. In the elliptic case this method work aswell, provided the total difference is
0 =
∮dNx
N∑i=1
1
xi(1− qxi∂i)
∑k∈Z
xitkN
z − q′kxi
∏j 6=i
Θq′
(txjxi
)Θq′
(xjxi
) ∆(q,q′,t)ell (x)
∼∼∮dNx∆
(q,q′,t)ell (x)
N∏j=1
Θq′
(txjz
)Θq′
(xjz
) + t2N−1q
N∏j=1
Θq′( qxjtz
)Θq′
( qxjz
) − QN (z)
. (74)
The resulting equation is, of course the same as Eq. (71). The meaning of the identity (74) in the ellipticmatrix model is the same as in the (q, t)-matrix model: it provides the recurrence relations for the correlators ofarbitrary symmetric functions of xi. It would be interesting to obtain the factorization formulas for the averagesin this model similar to those for the averages of (generalized) Macdonald polynomials in the (q, t)-model. Letus also mention that in the Nekrasov-Shatashvili limit Eq. (74) reduces to the quantum spectral curve of theXYZ spin chain, to the Seiberg-Witten integrable system corresponding to the 6d gauge theory.
This concludes our brief tour into the realm of elliptic matrix models and elliptic DIM algebras. The mostimportant lesson to learn here is that the DIM description indeed seems to be universal: the elliptic case isalmost literally the same as the trigonometric one.
4 Conclusions and further directions
We have worked out the connection between a large class of network matrix models associated with toric diagramsand the DIM algebra. The algebra provides a unified description of the symmetry behind all such matrix modelsgiving rise to qq-characters, generalized polynomials and Ward identities.
Application of the algebraic description to a matrix model such as (4) requires:
(i) identification of a particular free field representation of the appropriate DIM associated with the givenmodel,
(ii) building explicit expressions for the screening operators expressed as integrals of screening currents S(x),
(iii) constructing the symmetry generators (generalized stress-tensors) T (z) for which the screening operatorsare the centralizers,
(iv) representing the correlators of screening currents as Vandermonde measures and stress-tensor insertions asqq-characters which can be converted into the action of differential/difference operators. This step relieson the T -S OPE, which should be nonsingular up to a total difference, and the S-S OPE, which shouldgive the desired version of the Vandermonde determinant.
Schematically, one should have
T (z)S(x) = Regular(z, x) + (1− qx∂x) Singular(z/x)
S(x1)S(x2) = f(x1/x2) : S(x1)S(x2) : (75)
15
and the function f(x) defines the Vandermonde factor through ∆(x) =∏i6=j f(xi/xj). For the concrete examples
of OPEs like (75) see Eqs. (18), (21).It is still unclear how to separate the contributions of screening currents and vertex operators in the network
matrix model formalism since both objects are packed into a single intertwiner/topological vertex. Probably, thetechnical answer to this question should depend on the “star-chain” duality for conformal blocks.
This procedure is supposed to associate a D-module structure with each particular network matrix model or,what is the same, with representation of DIM. A non-trivial feature of actual construction, already seen in (9)and (17) is that the stress tensors are actually build from roots of algebra, while the screening operators fromCartan generators of DIM, which is somewhat against a naive intuition coming from their realization as powers∂φ and
∮e±φ in the simplest free field conformal theories. General understanding of this phenomena includes
relation between the Sugawara construction and the DIM comultiplication and between the screening chargesand the action of the Weyl group. Remarkably, the Weyl group of elliptic DIM should be the elliptic DAHA, ofwhich the elliptic Macdonald functions explicitly provided by formulas like (8) in elliptic matrix model (57), areeigenfunctions.
An interesting question here is interpretation of the BPZ equations [46] for such insertions as the Baxterequations for symmetric functions of Macdonald family, especially in elliptic case, where there exist alternativeapproaches [49].
ρu(x+(z))
u
v
−uv
=
(ρ/−uv ⊗ ρ
|v)∆(x+(z))
u
v
−uv
a)ρ(2)u1,u2(x
+(z))
=
(ρ/−u1v ⊗ ρ
|v ⊗ ρu2)∆
2(x+(z))
=
(ρ/−u1v ⊗ ρ
\−u2v)∆(x+(z))
u1
u2
u1
u2
u1
u2
b)
Figure 2: Topological vertex as the intertwiner of DIM representations. a) The action of the generator x+(z) on the level one Fockrepresentation ρu sitting on the horizontal leg of the topological vertex (denoted by the dashed line) is the same as its action on
the product of two representations — the “vertical” ρ|v and “diagonal” ρ
/−uv. b) Appropriate contraction of two intertwiners is
also an intertwiner. This gives the vertex operator of the corresponding conformal field theory with deformed Virasoro symmetry,corresponding to a single vertical brane in Fig. 1.
At the level of network matrix model (3) DIM symmetry generators act on any section which cuts M edgesto separate the diagram into disconnected parts (see Fig. 2). They act as (M − 1)-th coproduct of the originalDIM generators. As was shown in [33], topological vertices are intertwiners of DIM representations, i.e. theaction on one of the legs is equal to the action on two others — this allows to pull the generator through thevertex (Fig. 2 a)). Moreover, contraction of the legs is consistent with this procedure (Fig. 2 b)). In the resultDIM generators can be pulled from the original section to the right of the diagram, where negative modesannihilate the Fock vacuum, or to the left, where the positive modes act trivially. This provides the constraintson the matrix model averages (or, equivalently, gives the qq-characters), which are the 5d generalization of theconstraints obtained in [19]. Fig. 2 actually gives the uplift of the setup considered in [19] to the level of thetopological string (or network matrix model). Further developments of this approach and its applications tocompactified toric diagrams will be reported elsewhere.
Network matrix model is naturally built from the Seiberg-Witten integrable system — which is a spin chain inthe simplest cases [50, 40]. The network is the tropicalization of its spectral curve, and the vertical and horizontalbranes encode the rank and number of chain sites respectively. The structure of intertwiners/R-matrices forming anetwork can be understood as a lift of ordinary trigonometric R-matrices similar to the tetrahedron equation [51].This will give the connection between the algebraic and integrable parts of the story [31].
After the basic structure of network matrix model constraints is understood, we face a multitude of differentpaths, each one worth following. First of all, since DIM algebras involves double affinization of any Lie algebra(we have only considered gl1 case) it can be applied to the affine algebra gl1. This should provide a triply affine
algebra Uq,t,t(gl1) with three parameters. In this notation it seems appropriate to name this crucially important
16
structure the Pagoda Algebra. This algebra should have remarkable properties, one of which is the presence ofan SL(3,Z) automorphism group [52], corresponding to the automorphisms of the compactification torus T3.
In the second part of this paper we have considered elliptic DIM algebra, corresponding to 6d gauge theorywith matter content given by a linear quiver. The elliptization of the triply affine Pagoda algebra should,therefore, describe the 6d gauge theory with adjoint matter, the most mysterious of all Seiberg-Witten systems,corresponding to double-elliptic integrable systems and affine elliptic Selberg integrals. However, even withoutextra deformations, already the case of elliptic DIM poses interesting questions.
To summarize, the main idea of this paper is that DIM provides a functor, which lifts the picture — anetwork — to formulas made out of Nekrasov functions, 3d partitions or topological vertices. In other words,the input is a tropical spectral curve (associated with the underlying Seiberg-Witten integrable system) and theoutput is the partition function of the associated topological string theory, which is provided by one and thesame universal procedure. At the algebraic level the input should be the algebra gl1 which, treated as gl∞ orW1+∞, incorporates various gln’s, and the output is described by the Pagoda algebra, which still needs to befully investigated.
Acknowledgements
We are grateful to Prof. H. Kanno for remarkable hospitality at Nagoya University at the last stage of this project.We are deeply indebted to H. Awata, H. Kanno, Y. Ohkubo and V. Pestun for lecturing us on various aspects ofthe DIM symmetry and its applications and to T. Matsumoto and Yu. Matsuo for encouraging comments.
Our work is partly supported by grants 15-31-20832-Mol-a-ved (A.Mor.), 15-31-20484-Mol-a-ved (Y.Z.),by RFBR grants 16-01-00291 (A.Mir.) and 16-02-01021 (A.Mor. and Y.Z.), by joint grants 15-51-50034-YaF,15-51-52031-NSC-a, 16-51-53034-GFEN, by the Brazilian National Counsel of Scientific and TechnologicalDevelopment (A.Mor.).
References
[1] A. Mironov, A. Morozov and Y. Zenkevich, arXiv:1603.00304
[2] A. Morozov, Phys.Usp.(UFN) 35 (1992) 671-714; 37 (1994) 1, hep-th/9303139; hep-th/9502091; hep-th/0502010A. Mironov, Int.J.Mod.Phys. A9 (1994) 4355, hep-th/9312212; Phys.Part.Nucl. 33 (2002) 537; hep-th/9409190
[3] A. Marshakov, A. Mironov and A. Morozov, Phys. Lett. B265 (1991) 99S. Kharchev, A. Marshakov, A. Mironov, A. Morozov and S. Pakuliak, Nucl. Phys. B404 (1993) 717-750,hep-th/9208044R. Dijkgraaf and C. Vafa, arXiv:0909.2453
[4] B. Feigin and D. Fuks, Funct. Anal. Appl. 16 (1982) 114-126 (Funkt. Anal. Pril. 16 (1982) 47-63)Vl. Dotsenko and V. Fateev, Nucl. Phys. B240 (1984) 312-348H. Itoyama, K. Maruyoshi and T. Oota, Prog. Theor. Phys. 123 (2010) 957-987, arXiv:0911.4244T. Eguchi and K. Maruyoshi, arXiv:0911.4797; arXiv:1006.0828R. Schiappa and N. Wyllard, arXiv:0911.5337A. Mironov, A. Morozov and S. Shakirov, JHEP 1002 (2010) 030, arXiv:0911.5721; Int. J. Mod. Phys.A25 (2010) 3173, arXiv:1001.0563; J. Phys. A44 (2011) 085401, arXiv:1010.1734; JHEP 1103 (2011) 102,arXiv:1011.3481; Int. J. Mod. Phys. A27 (2012) 1230001, arXiv:1011.5629H. Itoyama and T. Oota, Nucl. Phys. B838 (2010) 298-330, arXiv:1003.2929A. Mironov, A. Morozov, and And. Morozov, Nucl. Phys. B843 (2011) 534, arXiv:1003.5752
[5] D. Galakhov, A. Mironov and A. Morozov, ZhETF 147 (2015) 623-663 (JETP 120 (2015) 623-663),arXiv:1410.8482
[6] M. Aganagic, N. Haouzi, C. Kozcaz and S. Shakirov, arXiv:1309.1687M. Aganagic, N. Haouzi and S. Shakirov, arXiv:1403.3657M. Aganagic and N. Haouzi, arXiv:1506.04183
17
[7] P. Sulkowski and A. Klemm, Nucl. Phys. B819 (2009) 400-430, arXiv:0810.4944P. Sulkowski, JHEP 04 (2010) 063, arXiv:0912.5476
[8] A. Iqbal, N. Nekrasov, A. Okounkov and C. Vafa, JHEP 0804 (2008) 011, hep-th/0312022
[9] A. Iqbal, C. Kozcaz and S. T. Yau, arXiv:1511.00458F. Nieri, arXiv:1511.00574
[10] A. Morozov and A. Smirnov, Lett. Math. Phys. 104 (2014) 585, arXiv:1307.2576S. Mironov, An. Morozov and Y. Zenkevich, JETP Lett. 99 (2014) 109, arXiv:1312.5732Y. Ohkubo, arXiv:1404.5401B. Feigin, M. Jimbo, T. Miwa and E. Mukhin, arXiv:1502.07194
[11] Y. Zenkevich, JHEP 1505 (2015) 131, arXiv:1412.8592A. Morozov and Y. Zenkevich, JHEP 1602 (2016) 098, arXiv:1510.01896A. Mironov, A. Morozov, Y. Zenkevich, arXiv:1512.06701
[12] E. Frenkel and N. Reshetikhin, Recent Developments in Quantum Affine Algebras and related topics, Cont.Math. 248 (1999) 163-205, math/9810055
[13] J. Ding, K. Iohara, Lett. Math. Phys. 41 (1997) 181–193, q-alg/9608002K. Miki, J. Math. Phys. 48 (2007) 123520
[14] B. Feigin, E. Feigin, M. Jimbo, T. Miwa and E. Mukhin, Kyoto J. Math. 51 (2011) 337–364, arXiv:1002.3100;365–392, arXiv:1002.3113
[15] H. Awata, B. Feigin, A. Hoshino, M. Kanai, J. Shiraishi and S. Yanagida, arXiv:1106.4088S. Kanno, Y. Matsuo and S. Shiba, Phys. Rev. D84 (2011) 026007, arXiv:1105.1667S. Kanno, Y. Matsuo and H. Zhang, arXiv:1207.5658; arXiv:1306.1523
[16] N. Nekrasov and V. Pestun, arXiv:1211.2240N. Nekrasov, S. Shatashvili and V. Pestun, arXiv:1312.6689
[17] N. Nekrasov, arXiv:1512.05388
[18] T. Kimura and V. Pestun, arXiv:1512.08533
[19] J.-E. Bourgine, Y. Matsuo and H. Zhang, arXiv:1512.02492
[20] A. Mironov and A. Morozov, Phys. Lett. B252 (1990) 47-52H. Itoyama and Y. Matsuo, Phys. Lett. B255 (1991) 202
[21] A.A. Migdal, Phys. Rep. 102 (1983) 199J. Ambjørn, J. Jurkiewicz and Yu. Makeenko, Phys. Lett. B251 (1990) 517F. David, Mod. Phys. Lett. A5 (1990) 1019J. Ambjørn and Yu. Makeenko, Mod. Phys. Lett. A5 (1990) 1753A.Alexandrov, A.Mironov, A.Morozov, Int.J.Mod.Phys. A19 (2004) 4127, hep-th/0310113; Teor.Mat.Fiz.150 (2007) 179-192, hep-th/0605171; Physica D235 (2007) 126-167, hep-th/0608228; JHEP 12 (2009) 053,arXiv:0906.3305A.Alexandrov, A.Mironov, A.Morozov, P.Putrov, Int.J.Mod.Phys. A24 (2009) 4939-4998, arXiv:0811.2825A.Mironov, A.Morozov, A.Popolitov, Sh.Shakirov, Theor.Math.Phys. 171 (2012) 505-522, arXiv:1103.5470Y. Zenkevich, arXiv:1507.00519
[22] Yu. Makeenko, A. Marshakov, A. Mironov and A. Morozov, Nucl. Phys. B356 (1991) 574
[23] A. Iqbal, hep-th/0207114M. Aganagic, A. Klemm, M. Marino and C. Vafa, Commun. Math. Phys. 254 (2005) 425 hep-th/0305132M. Taki, JHEP 0803 (2008) 048, arXiv:0710.1776H. Awata and H. Kanno, JHEP 0505 (2005) 039, hep-th/0502061; Int. J. Mod. Phys. A24 (2009) 2253,arXiv:0805.0191
[24] A. Iqbal, C. Kozcaz and C. Vafa, JHEP 0910 (2009) 069, hep-th/0701156
[25] I.P. Goulden and A. Rattan, Trans. Amer. Math. Soc., 359, (2007) 3669-3685;P. Biane, Lecture Notes in Math., 1815, (2003), 185-200
18
[26] A. Alexandrov, A. Mironov and A. Morozov, Int. J. Mod. Phys. A21 (2006) 2481-2518, hep-th/0412099;Fortsch.Phys. 53 (2005) 512-521, hep-th/0412205
[27] N. Seiberg and E. Witten, Nucl. Phys. B426 (1994) 19-52, hep-th/9407087; Nucl. Phys. B431 (1994)484-550, hep-th/9408099
[28] A.Gorsky, I.Krichever, A.Marshakov, A.Mironov, A.Morozov, Phys.Lett. B355 (1995) 466, hep-th/9505035R. Donagi and E. Witten, Nucl. Phys. B460 (1996) 299-334, hep-th/9510101
[29] H. W. Braden, A. Marshakov, A. Mironov, A. Morozov, Nucl. Phys. B573 (2000) 553–572, hep-th/9906240A. Mironov and A. Morozov, Phys. Lett. B475 (2000) 71-76, hep-th/9912088; hep-th/0001168G. Aminov, A. Mironov, A. Morozov and A. Zotov, Phys. Lett. B726 (2013) 802, arXiv:1307.1465G. Aminov, H. W. Braden, A. Mironov, A. Morozov and A. Zotov, JHEP 1501 (2015) 033 arXiv:1410.0698
[30] N. Guay, Adv. Math. 211 (2007) 436484D. Maulik and A. Okounkov, arXiv:1211.1287N. Arbesfeld and O. Schiffmann, arXiv:1209.0429O. Schiffmann and E. Vasserot, Publ. Math. Inst. Hautes Etudes Sci. 118 (2013) 213342, arXiv:1202.2756A. Smirnov, arXiv:1302.0799, arXiv:1404.5304A. Tsymbaliuk, arXiv:1404.5240R.-D. Zhu and Y. Matsuo, Prog. Theor. Exp. Phys. (2015) 093A01, arXiv:1504.04150M. Fukuda, S. Nakamura, Y. Matsuo and R.-D. Zhu, arXiv:1509.01000T. Prochazka, arXiv:1512.07178M. Bernshtein and A. Tsymbaliuk, arXiv:1512.09109
[31] B. Feigin, M. Jimbo, T. Miwa and E. Mukhin, Kyoto J. Math. 52, no. 3 (2012), 621-659, arXiv:1110.5310;arXiv:1204.5378; arXiv:1309.2147; arXiv:1603.02765
[32] B. Feigin, K. Hashizume, A. Hoshino, J. Shiraishi, S. Yanagida, J. Math. Phys. 50 (2009) 095215,arXiv:0904.2291
[33] H. Awata, B. Feigin and J. Shiraishi, arXiv:1112.6074
[34] M. Wakimoto, Commun. Math. Phys. 104 (1986) 605-609J. Wess and B. Zumino, Phys. Lett. B37 (1971) 95S. Novikov, UMN, 37 (1982) 37E. Witten, Comm. Math. Phys. 92 (1984) 455A. Gerasimov, A. Marshakov, A. Morozov, M. Olshanetsky and S. Shatashvili, Int. J. Mod. Phys. A5 (1990)2495-2589, DOI: 10.1142/S0217751X9000115XB. Feigin and E. Frenkel Phys. Lett. B246 (1990) 75-81
[35] H. Awata, Y. Matsuo, S. Odake and J. Shiraishi, Phys. Lett. B 347 (1995) 49 [hep-th/9411053]H. Awata, Y. Matsuo, S. Odake and J. Shiraishi, Soryushiron Kenkyu 91 (1995) A69-A75, hep-th/9503028;Nucl.Phys. B449 (1995) 347-374, hep-th/9503043
[36] H. Awata, S. Odake and J. Shiraishi, Commun. Math. Phys. 179 (1996) 647, q-alg/9506006
[37] S. Kharchev, A. Marshakov, A. Mironov and A. Morozov, Nucl. Phys. B397 (1993) 339-378, hep-th/9203043
[38] A. Marshakov, A. Mironov and A. Morozov, J. Geom. Phys. 61 (2011) 1203-1222, arXiv:1011.4491
[39] L. Bao, E. Pomoni, M. Taki and F. Yagi, JHEP 1204 (2012) 105, arXiv:1112.5228
[40] E. Mukhin, V. Tarasov and A. Varchenko, math/0510364; Adv. Math. 218 (2008) 216-265, math/0605172A. Mironov, A. Morozov, Y. Zenkevich and A. Zotov, JETP Lett. 97 (2013) 45, arXiv:1204.0913A. Mironov, A. Morozov, B. Runov, Y. Zenkevich and A. Zotov, Lett. Math. Phys. 103 (2013) 299,arXiv:1206.6349; JHEP 1312 (2013) 034, arXiv:1307.1502
[41] H. Awata and Y. Yamada, JHEP 1001 (2010) 125, arXiv:0910.4431; Prog. Theor. Phys. 124 (2010) 227,arXiv:1004.5122S. Yanagida, arXiv:1005.0216H. Awata, H. Fuji, H. Kanno, M. Manabe and Y. Yamada, Adv. Theor. Math. Phys. 16 (2012) no.3, 725[arXiv:1008.0574 [hep-th]].
19
A. Mironov, A. Morozov, S. Shakirov and A. Smirnov, Nucl. Phys. B855 (2012) 128, arXiv:1105.0948F. Nieri, S. Pasquetti, F. Passerini and A. Torrielli, arXiv:1312.1294H. Itoyama, T.Oota and R. Yoshioka, arXiv:1408.4216, arXiv:1602.01209A. Nedelin and M. Zabzine, arXiv:1511.03471R. Yoshioka, arXiv:1512.01084Y. Ohkubo, H. Awata and H. Fujino, arXiv:1512.08016
[42] A.Gerasimov, A.Marshakov, A.Morozov, Phys.Lett. B236 (1990) 269, DOI: 10.1016/0370-2693(90)90980-KA. Marshakov and A. Morozov Nucl.Phys. B339 (1990) 79-94, DOI: 10.1016/0550-3213(90)90534-K
[43] L. Alday, D. Gaiotto and Y. Tachikawa, Lett. Math. Phys. 91 (2010) 167–197, arXiv:0906.3219N. Wyllard, JHEP 0911 (2009) 002, arXiv:0907.2189A. Mironov and A. Morozov, Nucl. Phys. B825 (2009) 1–37, arXiv:0908.2569
[44] G.Moore, N.Nekrasov, S.Shatashvili, Nucl.Phys. B534 (1998) 549-611, hep-th/9711108; hep-th/9801061A.Losev, N.Nekrasov, S.Shatashvili, Comm.Math.Phys. 209 (2000) 97-121, hep-th/9712241; ibid. 77-95,hep-th/9803265
[45] N. Nekrasov, Adv. Theor. Math. Phys. 7 (2004) 831-864, hep-th/0206161R. Flume and R. Pogossian, Int. J. Mod. Phys. A18 (2003) 2541N. Nekrasov and A. Okounkov, hep-th/0306238
[46] A. Belavin, A. Polyakov and A. Zamolodchikov, Nucl. Phys. B241 (1984) 333-380A. Zamolodchikov, Al. Zamolodchikov, Conformal field theory and critical phenomena in 2d systems, 2009L. Alvarez-Gaume, Helvetica Physica Acta 64 (1991) 361P. Di Francesco, P. Mathieu and D. Senechal, Conformal Field Theory, Springer, 1996A.Mironov, S.Mironov, A.Morozov, An.Morozov, Theor.Math.Phys. 165 (2010) 1662-1698, arXiv:0908.2064
[47] V.A.Alba, V.A.Fateev, A.V.Litvinov, G.M.Tarnopolsky, Lett.Math.Phys. 98 (2011) 33-64, arXiv:1012.1312A. Belavin and V. Belavin, Nucl.Phys. B850 (2011) 199-213, arXiv:1102.0343Y. Matsuo, C. Rim and H. Zhang, arXiv:1405.3141
[48] V. Fateev and A. Litvinov, JHEP 1201 (2012) 051, arXiv:1109.4042
[49] A. V. Litvinov, JHEP 1311 (2013) 155, arXiv:1307.8094M. N. Alfimov and A. V. Litvinov, JHEP 1502 (2015) 150, arXiv:1411.3313G. Bonelli, A. Sciarappa, A. Tanzini and P. Vasko, JHEP 7 (2014) 141, arXiv:1403.6454; arXiv:1505.07116P. Koroteev and A. Sciarappa, arXiv:1510.00972; arXiv:1601.08238
[50] A. Gorsky, A. Marshakov, A. Mironov, A. Morozov, Phys. Lett. B380 (1996) 75-80, arXiv:hep-th/9603140;hep-th/9604078A. Gorsky, S. Gukov and A. Mironov, Nucl. Phys. B518 (1998) 689, hep-th/9710239; Nucl. Phys. B517(1998) 409-461, hep-th/9707120A. Marshakov and A. Mironov, Nucl. Phys. B518 (1998) 59-91, hep-th/9711156
[51] V. V. Bazhanov and S. M. Sergeev, J. Phys. A 39 (2006) 3295 doi:10.1088/0305-4470/39/13/009 [hep-th/0509181].
[52] G. Lockhart and C. Vafa, arXiv:1210.5909 [hep-th].K. Shabbir and K. Shabbir, Eur. Phys. J. C 76 (2016) no.3, 148 [arXiv:1510.03332 [hep-th]].
20
![dissipation and density modulation - arXivIn medical applications [BMTY02] each diffeomorphisms represents a particular anatomic configuration of an anatomic reference structures](https://img.pdfslide.us/doc/110x75/5fde819c6e75176c721ae707/dissipation-and-density-modulation-arxiv-in-medical-applications-bmty02-each.jpg)
