Discriminants and motivic integration

Oscar Kivinen, Alexei Oblomkov, Dimitri Wyss · 2025

Read Discriminants and motivic integration by Oscar Kivinen, Alexei Oblomkov, Dimitri Wyss (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-18449.txt.

Introduction

Oscar Kivinen, Alexei Oblomkov, and Dimitri Wyss present a 2025 arXiv paper in algebraic geometry, on invariants of a plane curve singularity. Using motivic integration over symmetric powers of a deformed curve, they study two kinds of discriminant integrals. One recovers the motivic classes of the principal Hilbert schemes of points on the curve. The other gives the plethystic exponential of the motivic Igusa zeta function, a result that also holds in higher dimensions. Combined with work of Gorsky and Némethi, the integrals gain an interpretation through knot Floer homology.

Catalog record

CREATOROscar Kivinen, Alexei Oblomkov, Dimitri Wyss
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE63 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-18449

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Discriminants and motivic integration Oscar Kivinen, Alexei Oblomkov, Dimitri Wyss Abstract . We study invariants of a plane cuve singularity ( f , 0 ) (f,0) coming from motivic integration on symmetric powers of a formal deformation of f f . We show that a natural discriminant integral recovers the motivic classes of the principal Hilbert schemes of points on f f , while the orbifold integral gives the plethystic exponential of the motivic Igusa zeta function of f f . The latter result also holds in higher dimemsions. Combined with results of Gorsky and Némethi we obtain an interpretation of the discriminant integrals in terms of knot Floer homology, which is reminiscent of the relation between the cohomology of contact loci and fixed point Floer homology proven by de la Bodega and Poza. 1. Introduction Consider a nonconstant polynomial map f:\mathbb{A}^{m}\to\mathbb{A}^{1} over k=\mathbb{C} defining an isolated hypersurface singularity at 0\in\mathbb{A}^{m} . A fundamental invariant of (f,0) is its Milnor fiber F_{0} , a compact smooth manifold with boundary depending only on the singularity at 0 [ 32 ] . Its singular cohomology H^{*}(F_{0},\mathbb{Q}) admits a monodromy endomorphism M , whose zeta function \zeta_{f,0}(T)=\prod_{i\geq i}\det(1-TM\ |\ H^{i}(F_{0},\mathbb{Q}))^{(-1)^{i+1}} can be computed purely algebraically by a remarkable result of Denef and Loeser [ 17 ] : For any n\geq 1 let X^{0}_{f,n} denote the n -th restricted contact locus of f . It is defined as X^{0}_{f,n}=\{x\in\mathbb{A}^{m}(k[[t]]/t^{n+1})_{0}\ |\ f(x)=t^{n}\in k[[t]]/t^{n+1}\}, where \mathbb{A}^{m}(k[[t]]/t^{n+1})_{0} denotes the locus of n -jets reducing to 0 modulo t . Then the Euler characteristics \chi(X^{0}_{f,n}) determine \zeta_{f,0}(T) via the formula (1.1) \zeta_{f,0}(T)=\exp\left(\sum_{n\geq 1}\frac{\chi(X^{0}_{f,n})}{n}T^{n}\right). While the original proof of ( 1.1 ) proceeds by computing both sides independently on an embedded resolution of f , there exists now a more conceptual interpretation [ 33 , 27 ] as a Lefschetz fixed point formula on the analytic Milnor fiber. In a similar spirit, the goal of this paper is to study refinements of ( 1.1 ) coming from motivic integration. Concretely, consider the k[[t]] -scheme \mathcal{X}_{f}=\textup{Spec}(k[[t]][x_{1},\dots,x_{m}]/f-t), and associated with it for any n\geq 0 the relative symmetric power \textup{Sym}_{n}\mathcal{X}_{f}=\mathcal{X}_{f}\times_{k[[t]]}\dots\times_{k[[t]]}\mathcal{X}_{f}/S_{n} . In a nutshell our paper studies motivic classes in the localized Grothendieck ring \mathcal{C}_{k}=K_{0}(Var_{k})[\mathbb{L}^{-1},(1-\mathbb{L}^{i})^{-1}:i\geq 1] coming from Cluckers-Loeser motivic integrals [ 10 ] on \textup{Sym}_{n}\mathcal{X}_{f}^{0}(k[[t]])=\{x\in\textup{Sym}_{n}(\mathcal{X}_{f})(k[[t]])\ |\ x_{k}=n[0]\}. Our first main result is in the case m=2 , i.e. when (f,0) defines an isolated plane curve singularity, and is partially motivated by the second author’s work on the relation between Hilbert schemes and knot invariants [ 35 , 34 ] . Namely, since \mathcal{X}_{f|k((t))} is a smooth curve, the generic fiber of \textup{Sym}_{n}\mathcal{X}_{f} admits another model over \textup{Spec}(k[[t]]) given by the relative Hilbert scheme of points \mathrm{Hilb}_{n}(\mathcal{X}_{f}) . The special fiber \mathrm{Hilb}_{n}(\mathcal{X}_{f})_{k} is the classical Hilbert scheme \mathrm{Hilb}_{n}(f) of length n subschemes on the singular curve f=0 and we write \mathrm{Hilb}_{n}^{1}(f)_{0}\subset\mathrm{Hilb}_{n}(f) for the locally closed subscheme parametrizing principal subschemes supported at 0 . Theorem 1.1 ( 3.3 ) . For any n\geq 1 there exists a generically non-vanishing definable volume form |\omega_{n}^{gel}| on \textup{Sym}_{n}\mathcal{X}_{f}^{0} such that \int_{\textup{Sym}_{n}\mathcal{X}_{f}^{0}}|\omega^{gel}_{n}|=\mathbb{L}^{-n}[\mathrm{Hilb}_{n}^{1}(f)_{0}]\in\mathcal{C}_{k}. By Proposition 3.6 , the generating series P_{gel}(T)=\sum_{n\geq 0}\int_{\textup{Sym}_{n}\mathcal{X}_{f}^{0}}|\omega^{gel}_{n}|T^{n}, can be identified with the one-variable motivic Poincaré series of Campillo, Delgado and Gusein-Zade [ 7 , 8 ] . Since |\omega^{gel}_{n}| is generically non-vanishing, one can compute the Euler characteristic specialization of P_{gel}(T) by the trace formula for motivic Serre invariants, recovering an old result of Milnor [ 32 , Lemma 10.1] : Corollary 1.2 ( 3.9 ) . For any reduced plane curve singularity (f,0) with link L we have \zeta_{f,0}(T)=\frac{A_{L}(T)}{1-T}, where A_{L}(T) denotes the normalized Alexander polynomial of the link. As a second corollary, we combine Theorem 1.1 with results of Gorsky and Némethi [ 21 ] to obtain an interpretation of the motivic integrals \int_{\textup{Sym}_{n}\mathcal{X}_{f}^{0}}|\omega^{gel}_{n}| in terms of the knot Floer homology HFL^{-}(L,v) of the link L defined by the plane curve singularity (f,0) [ 36 , 37 ] . Corollary 1.3 ( 5.4 ) . We have an equality P_{gel}(T)=\sum_{v\in\mathbb{Z}^{r},d\in\mathbb{Z}}\dim HFL^{-}_{d}(L,v)\mathbb{L}^{d-2h(\underline{v})}T^{|v|}, where h(\underline{v})=\textup{codim}_{k[[x,y]]}J_{\underline{v}} as in Section 3.5 . It seems tempting to try to relate Corollary 1.3 to the arc-Floer conjecture of Budur-de Bobadilla-Lê-Nguyen [ 6 ] , now a theorem for plane curve singularities [ 14 ] , predicting an isomorphism between the compactly supported cohomology of the contact loci X^{0}_{f,n} and the fixed point Floer homology of the n -th iterate of the monodromy of F_{0} . Although we do not know any direct relation between P_{gel}(T) and the geometry of the spaces X^{0}_{f,n} at the moment, our second main result shows that a closely related integral on \textup{Sym}_{n}\mathcal{X}_{f}^{0} is determined by the motivic Igusa zeta function [ 16 ] Z_{f}(T)=\sum_{n\geq 1}[X^{0}_{f,n}]\mathbb{L}^{-mn}T^{n}. Theorem 1.4 ( 4.2 ) . For any non-constant f\in k[x_{1},\dots,x_{m}] there exists a natural orbifold form |\omega_{orb}|^{1/2} on \textup{Sym}_{n}\mathcal{X}^{0}_{f} such that \sum_{n\geq 0}\int_{\textup{Sym}_{n}\mathcal{X}^{0}_{f}}|\omega_{orb}|^{1/2}T^{n}=\textup{Exp}\left(\mathbb{L}^{-\frac{m-1}{2}}Z_{f}(\mathbb{L}^{-\frac{m-3}{2}}T)\right), where \textup{Exp}:T\mathcal{C}_{k}[\mathbb{L}^{1/2}][[T]]\to 1+T\mathcal{C}_{k}[\mathbb{L}^{1/2}][[T]] denotes the plethystic exponential. For m=3 the relative Hilbert scheme \mathrm{Hilb}_{n}(\mathcal{X}_{f}) provides a crepant resolution of the generic fiber of \textup{Sym}_{n}\mathcal{X}^{0}_{f} and a formula similar to Theorem 1.4 has appeared in [ 38 ] , see also Remark 4.3 . Finally, for m=2 , we can relate the two volume forms |\omega^{gel}_{n}| and |\omega_{orb}|^{1/2} on \textup{Sym}_{n}\mathcal{X}^{0}_{f} by means of the formula \mathbb{L}^{-\mathrm{ord}_{\Delta_{f}}/2}|\omega_{orb}|^{1/2}=|\omega^{gel}_{n}|, where \Delta_{f}\subset\textup{Sym}_{n}\mathcal{X}_{f} denotes the discriminant, see Proposition 4.5 . In particular, due to the factor \mathbb{L}^{-\mathrm{ord}_{\Delta_{f}}/2} , the series P_{gel}(T) does not seem to be determined by the classes [X^{0}_{f,n}] . Instead, there is a natural 1 -parameter family Q_{f}(s,T)=\sum_{n\geq 0}\int_{\textup{Sym}_{n}\mathcal{X}^{0}_{f}}|\Delta_{f}|^{s}|\omega_{orb}|^{1/2}T^{n}\in\mathcal{C}_{k}[\mathbb{L}^{1/2}][[T,\mathbb{L}^{-s}]], for some formal parameter s , such that (1.2) Q_{f}(1/2,T)=P_{gel}(T)\ \text{ and }\ Q_{f}(0,T)=\textup{Exp}\left(\mathbb{L}^{-\frac{1}{2}}Z_{f}(\mathbb{L}^{\frac{1}{2}}T)\right). A good understanding of Q_{f}(s,T) might give a new way of relating Z_{f}(T) to \zeta_{f,0}(T) , however, even for a smooth curve computing Q_{f}(s,T) is equivalent to computing the Igusa zeta functions of all classical discriminant polynomials, see Section 6 . The latter is an open problem as far as we know, and we hope to come back to it in future work. The paper is organized as follows: In Section 2 we recall the necessary background on Hilbert schemes and motivic integration. In particular, we discuss Haiman’s charts of the Hilbert scheme of \mathbb{A}^{2} which we use in Section 3 to construct the definable form |\omega_{n}^{gel}| and prove Theorem 1.4 . In Section 4 we discuss the orbifold formalism for symmetric powers, establish Theorem 1.4 and compare the orbifold form |\omega_{orb}|^{1/2} with |\omega_{n}^{gel}| . In Section 5 we discuss the relation with (symplectic) knot invariants, in particular Corollary 1.3 , and speculate about possible refinements and extensions of our results. Finally, in Section 6 we discuss explicit examples of the series Q_{f}(s,T) and its specializations. Acknowledgements: We warmly thank Javier de la Bodega-Aldama, Arthur Forey, Paolo Ghiggini, François Loeser, Johannes Nicaise, Simon Pepin Lehalleur and Ilaria Rossinelli for interesting discussions around the subjects of this paper.We are very grateful to Eugene Gorsky for helping us with the proof of Proposition 3.6 . A.O. would like to thank EPFL, Lausanne for hospitality during the visits in September 2023 and January 2024, many ideas in the paper were developed during these stays. O.K and D.W. were supported the by Swiss National Science Foundation [no. 196960]. O.K. was also supported by a Väisälä project grant of the Finnish Academy of Science and Letters. A.O

Read the complete text