Summation Formulas for Harmonic Maass Forms

Nikolaos Diamantis, Joshua Pimm · 2025

Read Summation Formulas for Harmonic Maass Forms by Nikolaos Diamantis, Joshua Pimm (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-2509-22607.txt.

Introduction

Nikolaos Diamantis and Joshua Pimm prove a summation formula for harmonic Maass forms with no growth limits. These forms generalize modular forms and appear in number theory and string physics. A prior paper by the same circle gave such a formula only for forms of polynomial growth. Using the theory of L-series attached to these forms, the new note removes that restriction entirely.

From the general formula, they deduce a summation formula for the partition function, which counts ways to write a number as a sum of smaller ones. The same theory yields a formula for classical modular forms. It also gives the asymptotics of a Riesz sum attached to a cusp form.

Catalog record

CREATORNikolaos Diamantis, Joshua Pimm
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-22607

Browse the full library · About the Tower of Babel

Full text

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

Summation Formulas for Harmonic Maass Forms Nikolaos Diamantis, Joshua Pimm Abstract . In a previous work [ BDG+25 ] , a summation formula for harmonic Maass forms of polynomial growth was established. In this note, we use the theory of L L -series of harmonic Maass forms to state and prove a summation formula for such forms without any restrictions on their growth. We deduce a summation formula for the partition function. We further employ the same theory to derive a result on classical modular forms, namely, a summation formula and the asymptotics of a Riesz sum attached to a cusp form. 1. Introduction Summation formulas and related analytic techniques have not featured prominently in the theory of harmonic Maass forms, which is surprising given the important role such techniques have played in the theory of classical modular forms and their applications. Part of the reason for this absence was the lack of a systematic theory of L -series attached to harmonic Maass forms. However, such a theory is now available (see [ DLRR23 ] ) and several foundational aspects of it have been explored. An application based on a summation formula was already given in the work [ DLRR23 ] where the L -series for harmonic Maass forms was introduced: it represented a contribution to the “harmonic lift” problem, via a summation formula that sheds light on the behaviour of the “holomorphic part” of a harmonic Maass form with a given “shadow”. More recently, a summation formula for a certain Riesz means [ BDG+25 ] was established, but only for harmonic Maass forms of polynomial growth. That was enough to obtain arithmetic applications for Hurwitz class numbers and for Fourier coefficients of negative half-integral weight Eisenstein series, but the class of harmonic Maass forms of polynomial growth is quite restricted. In this note, we establish a summation formula for harmonic Maass form without any constraints on the growth. This will be stated and proved in general form in Theorem 3.3 , but here we present only a special case to give the general idea. Specifically, let, for some n_{0}\in\mathbb{N}_{0} , f(z)=\sum_{n\geq-n_{0}}a(n)e^{2\pi inz} be a weakly holomorphic modular form of weight k\in\frac{1}{2}\mathbb{N} for \Gamma_{0}(N) . Let b(n) be the n -th Fourier coefficient of g(z):=(\sqrt{N}\,z)^{-k}f(-1/N\,z) and suppose that a(n),b(n)=O(e^{C_{f}\sqrt{n}}) , as n\to\infty , for some C_{f}>0 . We then have the following. Theorem 1.1 . For C>\max(C^{2}_{f}\sqrt{N}/(8\pi),2\pi n_{0}/\sqrt{N}) and for all X>0 , we have \hskip-5.69054pt\sum_{n\geq-n_{0}}a(n)K_{0}\left(2\sqrt{D_{n}(C+\sqrt{N}X)}\right)\\ =i^{k}C^{\frac{k}{2}}\sum_{n\geq-n_{0}}b(n)\left(\sqrt{N}X+D_{n}\right)^{-\frac{k}{2}}K_{k}\left(2\sqrt{C(D_{n}+\sqrt{N}X)}\right) where D_{n}:=C+2\pi n/\sqrt{N} and K_{s}(z) is the modified Bessel function An immediate application of this is a summation formula for partition functions, which is stated in Corollary 3.4 . A second feature of the summation formula proved in [ BDG+25 ] is that it did not require the full use of the L -series defined in [ DLRR23 ] . Since it involved only polynomial growth harmonic Maass forms, the L -series attached to the specific type of harmonic Maass forms in [ SS22 ] sufficed. A natural question then is what kind of summation formulas for Riesz means we derive if we start from the more general L -series of [ DLRR23 ] having test functions as arguments. That means exploiting the freedom of selecting a suitable family of test functions, and in this note, we apply this principle to deduce a summation formula, a special case of which is the following: Theorem 1.2 . Let f be a holomorphic cusp form of weight k\in 2\mathbb{N} for \operatorname{SL}_{2}(\mathbb{Z}) with Fourier coefficients c(n) . Then, for \rho>1, and each X>0 we have \sum_{n<X/(2\pi)}\frac{c(n)e^{-2\pi n}}{(2\pi n)^{\frac{k-1}{2}}}\left(X-2\pi n\right)^{\rho}=i^{-k}X^{\rho}L_{f}\left(\psi_{\frac{k-1}{2}}\right)\\ +\frac{\Gamma(\rho+1)}{2\pi i^{k+1}\Gamma(k)}\sum_{n>0}\frac{c(n)}{e^{2\pi n}}\int_{(-\varepsilon)}X^{s+\rho}\frac{\Gamma\left(\frac{k+1}{2}-s\right)\Gamma(s)}{\Gamma(s+\rho+1)}M\left(\frac{k-1}{2}+s,k,2\pi n\right)ds (1.1) where M(a,b,z) is Kummer’s confluent hypergeometric function. The value L_{f}\left(\psi_{\frac{k-1}{2}}\right) on the right-hand side of ( 1.1 ) originates in the functional equation of L_{f}(\varphi) in the same way as the central critical value of the standard L -series of a holomorphic modular form. It would be interesting to see if there is a more intrinsic characterisation of such “critical values” of L_{f}(\varphi) . As an application of Theorem 1.2 , we deduce asymptotics for the weighted moment in the LHS of ( 1.1 ): Corollary 1.3 . With the assumptions and notation of Theorem 1.2 , we have, for \rho>1 and each X>0 , \sum_{n<X/(2\pi)}\frac{c(n)e^{-2\pi n}}{(2\pi n)^{\frac{k-1}{2}}}\left(1-\frac{2\pi n}{X}\right)^{\rho}=i^{-k}L\left(\psi_{\frac{k-1}{2}}\right)+O(X^{-\varepsilon}). (1.2) In the next section, we present the basic objects we will be studying. In Section 3 , we state and prove the summation formula for general harmonic Maass forms. In the final section, we prove the Riesz-type summation formula and use it to derive the application to weighted moments of Fourier coefficients of holomorphic cusp forms. 2. Preliminaries Let k\in\frac{1}{2}\mathbb{Z} and N\in\mathbb{N}. We write an element of the upper half-plane \mathbb{H} as z=x+iy , with x,y\in\mathbb{R} , y>0 . Further q:=e^{2\pi i\,z}. We choose the principal branch of the square-root throughout. The (Petersson) slash operator is defined as \displaystyle\left(f|_{k}\gamma\right)(\,z):=\begin{cases}(c\,z+d)^{-k}f(\gamma\,z)&\text{if }k\in\mathbb{Z},\\ \left(\frac{c}{d}\right)\varepsilon_{d}^{2k}(c\,z+d)^{-k}f(\gamma\,z)&\text{if }k\in\frac{1}{2}+\mathbb{Z},\end{cases}\quad \displaystyle\text{for}\,\,\gamma=\left(\begin{matrix}a&b\\ c&d\end{matrix}\right)\in\begin{cases}\operatorname{SL}_{2}(\mathbb{Z})&\text{if }k\in\mathbb{Z},\\ \Gamma_{0}(4)&k\in\frac{1}{2}+\mathbb{Z},\end{cases} where \left(\frac{c}{d}\right) denotes the Kronecker symbol, and \displaystyle\varepsilon_{d}:=\begin{cases}1&\text{if }d\equiv 1\pmod{4},\\ i&\text{if }d\equiv 3\pmod{4}.\end{cases} We define the action of w_{N}=\left(\begin{smallmatrix}0&-1/\sqrt{N}\\ \sqrt{N}&0\end{smallmatrix}\right) on a function f:\mathbb{H}\to\mathbb{C} as f|_{k}w_{N}(\,z):=N^{k/2}(N\,z)^{-k}f(-1/N\,z). (2.1) For \varphi:\mathbb{R}_{+}\to\mathbb{C}, we also define the function \varphi|_{k}w_{N} given by \varphi|_{k}w_{N}(x):=(Nx)^{-k}f(1/Nx)\qquad\text{for $x\in\mathbb{R}_{+}$} (2.2) and, for s\in\mathbb{C} , we set \varphi_{s}(x):=\varphi(x)x^{k-1}. (2.3) Definition 2.1 . Let k\in\frac{1}{2}\mathbb{Z} and N\in\mathbb{N} (with 4|N whenever k\in\frac{1}{2}+\mathbb{Z} ). Let \chi be a Dirichet character \mod N. We call a smooth function f\colon\mathbb{H}\to\mathbb{C} a harmonic Maass form of weight k and character \chi on \Gamma_{0}(N) , if it satisfies the following conditions: (1) For all \gamma\in\Gamma_{0}(N) , we have f|_{k}\gamma=\chi(\gamma)f . (2) We have \Delta_{k}f=0 , where \Delta_{k}:=-y^{2}\left(\frac{\partial^{2}}{\partial x^{2}}+\frac{\partial^{2}}{\partial y^{2}}\right)+iky\left(\frac{\partial}{\partial x}+i\frac{\partial}{\partial y}\right). (3) For each \sigma\in\operatorname{SL}_{2}(\mathbb{Z}) , there is a polynomial P_{\sigma} such that (f|_{k}\sigma)(z)-P_{\sigma}(q^{-1})=O(e^{-\varepsilon_{\sigma}y}) as y\to\infty , for some \varepsilon_{\sigma}>0 . We let H_{k}(N,\chi) denote the space of such functions. Each f\in H_{k}(N,\chi) has a Fourier expansion of the form f(\,z)=\sum_{n\geq-n_{0}}^{\infty}c^{+}(n)q^{n}+\sum_{n=1}^{\infty}c^{-}(n)\Gamma(1-k,4\pi ny)q^{-n}. (2.4) and similar expansions at all other cusps. For some C>0, we have c^{+}(n)=O(e^{C\sqrt{n}}) as n\to\infty , while c^{-}(n) have polynomial growth as n\to\infty (see [ BFOR17 , Sec. 5.2] ) If c^{-}(n)=0 for all n>0 and at all cusps, then f is called a weakly holomorphic modular form. Their space is denoted by M^{!}_{k}(N,\chi) . If, further c^{+}(0)=0 (at all cusps), f is called a weakly holomorphic cusp form. Their space is denoted by S^{!}_{k}(N,\chi) . We retrieve the space of weight k holomorphic modular (resp. cusp) forms for level N as the space M_{k}(N,\chi) (resp. S_{k}(N,\chi) ) of f\in M^{!}_{k}(N,\chi) such that c^{+}(n)=0 for all n<0 (resp. n\leq 0 ) (and similarly at the other cusps). Another important subspace of H_{k}(N,\chi) is the space H^{\prime}_{k}(N,\chi) of harmonic Maass forms such that for each \sigma\in\operatorname{SL}_{2}(\mathbb{Z}) , (f|_{k}\sigma)(z)-a_{\sigma}=O(e^{-\varepsilon_{\sigma}y}) as y\to\infty , for some a_{\sigma}\in\mathbb{C} and \varepsilon_{\sigma}>0 . The spaces H_{k}(N,\chi) and S_{2-k}(N,\bar{\chi}) are connected by the operator \xi_{k} mapping a f\in H_{k}(N,\chi) with expansion ( 2.4 ) to its “shadow” \xi_{k}f:=2iy^{k}\overline{\frac{\partial f}{\partial\bar{z}}}=-\sum_{n=1}^{\infty}\overline{c^{-}(n)}(4\pi n)^{1-k}q^{n}\in S_{2-k}(N,\bar{\chi}). (2.5) This, in particular, implies that H_{k}(N,\chi)=M^{!}_{k}(N,\chi),\qquad\text{if $k>2$.} (2.6) We now recall the definition of L -series of f\in H_{k}(N,\chi) given in [ DLRR23 ] . The domain \mathcal{F}_{f} we will be using in this is note consists of \varphi\in C(\mathbb{R},\mathbb{C}) such that the Laplace integral (\mathcal{L}\varphi)(s)=\int_{0}^{\infty}\varphi(x)e^{-sx}dx (resp. (\mathcal{L}\varphi_{2-k})(s) ) converges absolutely for all s with \operatorname{Re}(s)\geq-2\pi n_{0}

Read the complete text