Stable quotients and the holomorphic anomaly equation (Q1639644)

Let \(X\) be a (not necessarily compact) Calabi-Yau threefold and denote by \(Y\) its mirror. Genus 0 mirror symmetry states that the \(A\)-model data of genus 0 Gromov-Witten (GW) invariants of \(X\) is equivalent, via the mirror map, to the \(B\)-model data of periods of \(Y\). An extension to higher genus was proposed by \textit{M. Bershadsky} et al. [Nucl. Phys., B 405, No. 2--3, 279--304 (1993; Zbl 0908.58074)]: higher genus GW invariants should correspond to topological string amplitudes \(\mathcal{F}_g\), \(g\geq 1\), which satisfy some recursive partial differential equations called the holomorphic anomaly equations. For toric Calabi-Yau threefolds, in the \(B\)-model this is proven with the technique of the topological recursion by \textit{B. Eynard} and \textit{N. Orantin} [J. Phys. A, Math. Theor. 42, No. 29, Article ID 293001, 117 p. (2009; Zbl 1177.82049)]. The authors of the paper under review take a different approach. \textit{I. Ciocan-Fontanine} and \textit{B. Kim} [Adv. Math. 225, No. 6, 3022--3051 (2010; Zbl 1203.14014)] proved that for the quintic threefold, under application of the mirror map, the GW invariants correspond to the stable quotient invariants in all genus. Hence it is expected that the stable quotient invariants of \(X\) match the \(B\)-model invariants of \(Y\). This lead the authors of the present paper to prove the holomorphic anomaly equation directly by working with the moduli space of stable quotients. They develop a general formalism that applies to twisted theories on projective spaces (no longer necessarily Calabi-Yau). In this paper, they prove it for the local surface \(K\mathbb{P}^2\) given as the total space of the canonical bundle on \(\mathbb{P}^2\). In subsequent work, they prove it for the formal quintic. Motivated by this approach, Guo-Janda-Ruan announced a proof of the holomorphic anomaly equation for the quintic threefold. The methods used in the paper involve among others localization, stable graphs and I-functions. In the last section of the paper, the authors study the equivariant stable quotient theory of \(K\mathbb{P}^2\) and prove the holomorphic anomaly equation with insertions. Note the generalization of stable quotient invariants to quasimap invariants, which are defined for varieties arising from certain smooth projective GIT quotients. To learn more about these moduli spaces, the interested reader is referred to the works of Ciocan-Fontanine, Kim and their collaborators.