Quantum Serre theorem as a duality between quantum D-modules

Quantum Serre theorem as a duality between quantum -modules

Hiroshi Iritani Hiroshi Iritani, Department of Mathematics, Kyoto University, Kitashirakawa-Oiwake-cho, Sakyo-ku, Kyoto, 606-8502, Japan. iritani@math.kyoto-u.ac.jp Etienne Mann Université Montpellier 2, Institut de Mathématique et de Modélisation de Montpellier, UMR 5149, Case courier 051 Place Eugène Bataillon F-34 095 Montpellier CEDEX 5, France. etienne.mann@math.univ-montp2.fr http://www.math.univ-montp2.fr/ mann/index.html  and  Thierry Mignon Université Montpellier 2, Institut de Mathématique et de Modélisation de Montpellier, UMR 5149, Case courier 051 Place Eugène Bataillon F-34 095 Montpellier CEDEX 5, France. thierry.mignon@math.univ-montp2.fr http://www.math.univ-montp2.fr/ mignon/
Abstract.

We give an interpretation of quantum Serre theorem of Coates and Givental as a duality of twisted quantum -modules. This interpretation admits a non-equivariant limit, and we obtain a precise relationship among (1) the quantum -module of twisted by a convex vector bundle and the Euler class, (2) the quantum -module of the total space of the dual bundle , and (3) the quantum -module of a submanifold cut out by a regular section of .

When is the anticanonical line bundle , we identify these twisted quantum -modules with second structure connections with different parameters, which arise as Fourier-Laplace transforms of the quantum -module of . In this case, we show that the duality pairing is identified with Dubrovin’s second metric (intersection form).

Key words and phrases:
quantum cohomology, Gromov-Witten invariants, quantum differential equation, Fourier-Laplace transformation, quantum Serre, second structure connection, Frobenius manifold
2010 Mathematics Subject Classification:
14N35, 53D45, 14F10
Contents

1. Introduction

Genus-zero Gromov-Witten invariants of a smooth projective variety can be encoded in different mathematical objects: a generating function that satisfies some system of PDE (WDVV equations), an associative and commutative product called quantum product, the Lagrangian cone of Givental [Givental:symplectic] or in a meromorphic flat connection called quantum connection. These objects are all equivalent to each other; in this paper we focus on the realization of Gromov-Witten invariants as a meromorphic flat connection.

Encoding Gromov-Witten invariants in a meromorphic flat connection defines the notion of quantum -module [Givental:ICM], denoted by , that is a tuple consisting of a trivial holomorphic vector bundle over with fiber , a meromorphic flat connection on given by the quantum connection:

and a flat non-degenerate pairing on given by the Poicaré pairing (see Definition LABEL:def:quantumD-mod and Remark LABEL:rem:connection_z). These data may be viewed as a generalization of a variation of Hodge structure (see [Katzarkov-Pantev-Kontsevich-ncVHS]).

Quantum Serre theorem of Coates and Givental [Givental-Coates-2007-QRR, §10] describes a certain relationship between twisted Gromov-Witten invariants. The data of a twist is given by a pair of an invertible multiplicative characteristic class and a vector bundle over . Since twisted Gromov-Witten invariants satisfy properties similar to usual Gromov-Witten invariants, we can define twisted quantum product, twisted quantum -module and twisted Lagrangian cone associated to the twist . Let denote the characteristic class satisfying for any vector bundle . Quantum Serre theorem (at genus zero) gives the equality of the twisted Lagrangian cones:

(1.1)

Quantum Serre theorem of Coates and Givental was not stated as a duality. An observation in this paper is that this result can be restated as a duality between twisted quantum -modules:

Theorem 1.1 (see Theorem LABEL:thm:quantum,Serre,QDM for more precise statements).

There exists a (typically non-linear) map (see (LABEL:eq:f,map)) such that the following holds:

  1. The twisted quantum -modules and are dual to each other; the duality pairing is given by the Poincaré pairing.

  2. The map sending to is a morphism of quantum -modules.

Genus-zero twisted Gromov-Witten invariants were originally designed to compute Gromov-Witten invariants for Calabi-Yau hypersurfaces or non-compact local Calabi-Yau manifolds [Kontsevich:enumeration, Givental-Equivariant-GW, LocalMirrorSymmetry-Chiang-Klemm-Yau-Zaslow-99]. Suppose that is a convex vector bundle and is the equivariant Euler class . In this case, non-equivariant limits of -twisted Gromov-Witten invariants yield Gromov-Witten invariants of a regular section of and non-equivariant limits of -twisted Gromov-Witten invariants yield Gromov-Witten invariants for the total space . The original statement (1.1) of quantum Serre theorem does not admit a non-equivariant limit since the non-equivariant Euler class is not invertible. We see however that our restatement above passes to the non-equivariant limit as follows:

Corollary 1.2 (Theorem LABEL:thm:euler,quantum,Serre,QDM, Corollary LABEL:cor:Z).

Let be a convex vector bundle and let denote the (non-equivariant) Euler class. Let be the map given by and let denote the non-equivariant limit of the map of Theorem 1.1 in the case where . We have the following:

  1. The quantum -modules and are dual to each other.

  2. Let be the zero-locus of a regular section of and suppose that satisfies one of the conditions in Lemma LABEL:lem:cond_Z. Denote by the inclusion. Then the morphism factors through the ambient part quantum -module of as:

    (1.2)
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
332314
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description