Quantum D-modules of projective bundles and blowups
Abstract.
In this lecture series, we review the basics of quantum D-modules and Givental’s formalism. We then explain the projective bundle formula and the blowup formula, which are proved using Fourier transforms associated with rank-one torus actions. If time permits, we will also discuss a nonabelian version of the Fourier transform.
1 GW theory
One motivation for us comes from mirror symmetry. Let be a Fano variety (or more generally, log Calabi-Yau). Then we expect there to be a mirror , which is an LG model. Via some version of Saito theory (which may not actually be well-defined), we obtain a Gauss-Manin conjecture. This will correspond to the quantum D-module of , which is defined using Gromov-Witten theory.
Another motivation comes from the study of other areas of algebraic geometry1. Katzarkov-Kontsevich-Pantev-Yu define Hodge atoms which establish an irrationality criterion for varieties and was used to prove the irrationality of the cubic fourfold. Recently, an unreleased internal model at OpenAI proved the generalized Mukai conjecture using Gromov-Witten theory.
1.1 Quantum D-modules.
Let be a smooth projective variety and be a homogeneous basis with corresponding coordinates . Then we define the (big) quantum cohomology
where keeps track of effective curve classes, and here we are taking a graded completion of where
Example 1.1.
Let and . Then the curve classes are just non-negative integers, so has degree . In this case, .
Example 1.2.
Let have where and . In this case, we actually have
which lives in all even degrees.
Now let . The quantum product is defined by the formula
where is the Poincaré pairing and .
Remark 1.3.
We can define the quantum cohomology as a formal Frobenius manifold, but this is equivalent to the approach taken by the speaker, so we will skip it.
Definition 1.4.
We will now define the quantum D-module
The connection is known as the quantum connection and is defined by
Here, the first definition makes sense only for and the Euler vector field is defined by
Additionally, the grading operator is defined by . Finally, the pairing
is given by the formula
Example 1.5.
Let . Then we have
and
with the relevant connections.
Remarks 1.6.
- (1)
The object is not a D-module in the usual sense of D-modules.2
- (2)
The objects and contain the same information, namely that one can be recovered from the other.
- (3)
The ring is a quantization of the usual cohomology ring as a ring and is a quantization of as a graded -module.
We will now produce the fundamental solution to the quantum connection. This is given by the formula
for all .
Proposition 1.7.
We have
Moreover, and . We also have the asymptotics
and the property
1.2 Givental formalism.
Define
and the genus zero descendent GW potential
Now define the symplectic vector space
This is actually functorial, but we will skip the discussion of this. Then define the form
by the formula
This makes an infinite-dimensional symplectic vector space over .
There is a decomposition
where
Finally, we can define to be the graph of and refer to it as the Givental cone.
Theorem 1.8.
is an overruled Lagrangian cone; namely:
- (1)
is Lagrangian;
- (2)
It is a cone, namely that if and if ;
- (3)
For any , we have .
Example 1.9.
When , we see that the basis is dual to the basis . When we multiply by , we see that is not preserved.
Remark 1.10.
Let be a formal function on . Then the graph of is an overruled Lagrangian cone if and only if satisfies the string equation, dilaton equation, and topological recursion relations.
There is an informal description of the Lagrangian cone as
In particular, we see that
is an isomorphism intertwining and . This essentially explains all of genus zero Gromov-Witten theory.
| differential operator | hard | easy |
|---|---|---|
| description as -module | easy | hard |
For example, if we consider
we prove that the bottom arrow is an isomorphism, then go around the diagram to obtain the top arrow.
2 Equivariant GW theory
Now let be a smooth semiprojective (projective over affine) -variety, where is a torus. We can run the same formalism as before to obtain
This still intertwines and , but there are additional operators, called shift operators, which we will denote by and .
In this setting, we have the genus-zero potential
and therefore , , , and . Here, we will denote by coordinates corresponding to a homogeneous basis of and coordinates corresponding to a basis of (here this is a non-canonical isomorphism of -modules due to Goresky-Kottwitz-MacPherson).
Note here that we need to make the modification
and define
and the operator
We also can define
and the Givental cone . Note here that the localization occurs in the category of graded rings, so expressions of the form live in .
2.1 Shift operators.
Example 2.1.
Let carry the scaling action of . Note that has degree , so
and
Since the classical and quantum products coincide, and are easy to define. The operator is an endomorphism of given by the formula
Note that gives an isomorphism , and . In this setting, is given by
so the terms containing come from .
Remark 2.2.
From the example we can see that and do not depend on , but does.
Remark 2.3.
Later, we will use shift operators to lift Kirwan surjectivity
to obtain an isomorphism
More generally, we expect that the “completion”
is a quantization of the usual Kirwan map and that is a Kähler moduli space. with various limit points being GIT quotients or .
In the example, if we consider small quantum cohomology, we see that
Dependence on the dimension can be recovered by considering the structure as a -module.
Remark 2.4.
In symplectic topology, we can study quantum cohomology using the loop space
From the torus action on , we obtain a torus action on given by
There is also an action of on , where if is a group homomorphism, then
This action of is where the shift operators come from.
We will now actually introduce shift operators, which defines a -module structure on , where is an equivariant analogue of the monoid of effective curve classes.
Definition 2.5.
Define
to be the subgroup generated by vertical curve classes with respect to the fibration
where and by for any and .
Lemma 2.6.
The complex
is exact.
Sketch of proof.
By definition, we see that . Then we consider the diagram
where the vertical arrows are injective by definition and the bottom arrow is injective by the spectral sequence for . This implies the top arrow is injective, so it remains to show that the relevant image and kernel coincide. It is enough to show that for all lying on different fixed components and , then . It is clear that they are linked by a chain of -fixed rational curves. Note that if lie on the same fixed component, then they define the same homology class.
Now let . Then between and is a chain of Hirzebruch surfaces. For simplicity, suppose there are two surfaces and , both attaching to . If and denote fiber classes, we see that
□
Now let and set . Define
where the action is defined by . This can be viewed as some form of a “twisted” graph space and is the fiber product
There is an action of on given by the formula
Denote by and be the fibers at and . Then the action of on is given by
whereas the action on is given by
The identity map induces a nontrivial map
Here, we have to identify with itself by .
Now for any
there exists some such that and . Moreover, there exists a unique connected component whose normal bundle has only negative weights (using Morse theory for the real moment map, this is the top connected component of the fixed locus, which is the critical locus). Let be the section class associated to .
Definition 2.7.
For , define the shift operator
as follows: for any and , we have
Note that is -linear, but because is not linear, neither is , and in fact we have
Equivalently, for all , we have the commutator identity
Remark 2.8.
The map is not is not -linear, so we need to work over .
Definition 2.9.
Define
Proposition 2.10.
The shift operators give the structure of a -module.
2.2 Shift operators for Givental cones.
Define
where the additional factor of acts trivially. By the localization formula, we know
Definition 2.11.
For , define
as follows: for any fixed component and , we have
where we decompose and are the Chern roots of .
Proposition 2.12.
- (1)
We have ;
- (2)
commutes with commutators via
- (3)
The shift operators intertwine the fundamental solution:
Corollary 2.13.
- (1)
We have ;
- (2)
commutes with commutators via
3 Fourier transforms
There are two kinds of Fourier transforms, which were introduced by Iritani: continuous and discrete. Let be a smooth semi-projective -variety, where .
3.1 Discrete Fourier transform.
This is a quantization of the Kirwan map. Let be a -linearized ample line bundle on such that is acted on freely by and is smooth and projective. Then there is the Kirwan map
Conjecturally, this should quantize to
It will intertwine the shift operator with the Novikov variable and the equivariant parameter with the quantum connection .
Example 3.1.
Let with the diagonal action. Then the Kirwan map is
At the level of small quantum D-modules, we have
but we have a nontrivial shift operator
In the quotient, we have
where . Note here that whereas . Note here that
We may now define the map
More generally, we have
In this case, is an isomorphism. It is clear that it intertwines and . It is a bit less obvious that it intertwines and , but this is also not hard to check.
Note that has the action of for all , whereas has the action of .
Definition 3.2.
Let be the GIT chamber. Then we define
In progress work of Iritani-Sanda will show that .
Now we can define to be the graded completion of
Here, note that .
Conjecture 3.3 (Reduction conjecture, Iritani, Iritani-Sanda).
There exists , or equivalently, over , such that there exists and fitting into the diagram
such that on the left side is intertwined with and on the right side, and on the left side is intertwined with on the right side. Moreover, the bottom arrow is given by
Example 3.4.
Let with diagonal torus action. Then . Its Fourier trasnform is given by
Note that , so , and therefore
Applying again, we see that
On the other hand, we see that is killed by the Kirwan map, so in fact we see that
To recover the big I-function, just replace by .
Theorem 3.5.
The reduction conjecture is true for:
- •
Projective bundles , due to Iritani-Koto;
- •
(Possibly non-split) toric bundles , due to Koto;
- •
When and there exists a -fixed divisor such that acts on with weight and , due to Iritani. This plays a crucial role in the blowup formula.
3.2 Continuous Fourier transform.
This is a quantization of equivariant localization. Let be a connected component of the fixed locus. Then there exists
where is determined the action of on . Here, it is defined as a phase asymptotic approximation of the integral
Informally, define
Now consider the map
This intertwines with . Applying the classical idea of the Mellin transform, we can consider the integral
and turn the shift operator into a variable . Unfortunately, this integral is ill-defined because the integrand contains and .
Instead, we will consider a stationary phase approximation. Here, , where is a section. Then, we can rewrite the integral as
where
Here, if we write
then and . Finally, we define
Setting and (up to scaling by a rational number), then we have the stationary phase approximation
Note here that .
Definition 3.6.
We define
Here, we write , where . For all , there is corresponding to all of the critical points of .
Theorem 3.7 (Iritani).
For all there exists and a Fourier transform fitting into a commutative diagram
which intertwines and on the left with formal variables on the right and intertwines on the left with and on the right. Here, the are the critical points of .
Remark 3.8.
Iritani actually introduced
Example 3.9.
Again, consider with the diagonal action of . Then set . Note here that and . Then we see that
where when is odd and when is even. Note that this sends
Also, this intertwines and with and . Moreover, is intertwined with .
4 Projective bundle formula
Let be a vector bundle of rank over a smooth projective base . Assume that there exists a surjection , which ensures that is semi-projective. If this is not true, we can replace by for some . Then we see that
for some and .
The reduction conjecture predicts the existence of a Fourier transform
There is also a continuous Fourier transform
We will write and note that
where . We note that
Here, is associated to , where is the structure morphism.
Theorem 4.1 (Iritani-Koto).
There exists
for such that there exists
Here, . Moreover, intertwines the quantum connections and is linear over .
Sketch of proof.
The first step is to prove the reduction conjecture for . The second step is to prove that is an isomorphism. Once we have done this, we define a -module homomorphism
as
Unfortunately, this is linear only over , so we need a lift (here, and ) of . This is not hard to find, so we obtain a morphism
where . Now if we set and consider , we see that
where . Once we invert , we see that is an isomorphism. □
Remark 4.2.
For example, when is a point, then (here the big I-function is defined using light markings) via Birkhoff factorization. This defines as a mirror map.
Corollary 4.3.
As rings, we have an isomorphism
over . Moreover, the genus-zero Gromov-Witten invariants of can be recovered from the genus-zero invariants of and the total Chern class .
Theorem 4.4.
We have
where is the equivariant J-function of .
Sketch of proof.
Consider the short exact sequence
This gives us an inclusion , which descends to
which is cut out by a regular section of . We can now use quantum Lefschetz to relate and . Quantum Riemann-Roch relates this to , the work of Brown relates and , and finally quantum Riemann-Roch relates and . □
Remark 4.5.
To prove a mirror theorem for general non-split toric bundles, this strategy fails because is not convex in general. However, we can use virtual localization to reduce to the case of projective bundles.
5 Blowup formula
Let .
Theorem 5.1.
There exists a map
and a map
Here, is the codimension of and the maps are defined over .
Corollary 5.2.
We obtain a corresponding decomposition
over .
Theorem 5.3.
The genus-zero Gromov-Witten invariants of can be recovered from those of and , and from .
Proof.
Consider the master space with the action of . It has two GIT quotients and . There are three fixed components
so we have three continuous Fourier transforms and two conjectural discrete Fourier transforms
The first step of the proof is to show that contains essentially the same information as and that contains essentially the same information as . The next step is to prove that
is an isomorphism. Then we combine the Fourier transforms to construct and finally show that is an isomorphism. □
Example 5.4.
Let . Assume that and and set . Note that
where and . Using the J-function, we can compute the small quantum cohomology
where and . Doing some algebra, we can obtain an ad hoc blowup formula and deduce
On the other hand, an ad hoc version of the projective bundle formula viewing as an -bundle over gives us
Therefore, we have the following picture of :