On q q -Shuffle Relations for Multiple Eisenstein Series in Positive Characteristic

Ting-Wei Chang, Song-Yun Chen, Fei-Jun Huang, and Hung-Chun Tsui · 2025

Read On q q -Shuffle Relations for Multiple Eisenstein Series in Positive Characteristic by Ting-Wei Chang, Song-Yun Chen, Fei-Jun Huang, and Hung-Chun Tsui (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-2504-18879.txt.

Introduction

Ting-Wei Chang, Song-Yun Chen, Fei-Jun Huang, and Hung-Chun Tsui define a family of series in prime characteristic, in this 2025 note given in short form. The catalog title is garbled with funding text, so it follows the head of the text. Multiple Eisenstein series come in any rank, and known multiple zeta values appear as the constant terms of their expansions through multiple Goss sums. The authors prove the series obey the same q-shuffle relations as the zeta values, lifting them from values to functions.

Catalog record

CREATORTing-Wei Chang, Song-Yun Chen, Fei-Jun Huang, and Hung-Chun Tsui
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE61 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-18879

Browse the full library · About the Tower of Babel

Full text

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

On q q -Shuffle Relations for Multiple Eisenstein Series in Positive Characteristic Thanks: The first author is supported by the National Science and Technology Council grant no. 109-2115-M-007-017-MY5 and 113-2628-M-007-003. The latter three authors are supported by NSTC Grant Number 113-2628-M-007-004 Ting-Wei Chang, Song-Yun Chen, Fei-Jun Huang, and Hung-Chun Tsui Abstract . In this paper, we define the multiple Eisenstein series in positive characteristic, with Thakur’s multiple zeta values appearing as the “constant terms” of their expansions in terms of “multiple Goss sums”. We show that the multiple Eisenstein series satisfy the same q q -shuffle relations as the multiple zeta values do, thereby lifting the relations from “values” to “functions”. 2020 Mathematics Subject Classification Primary 11R58, 11M36; Secondary 11M32, 11M38 1. Introduction 1.1. The Classical Results Let \mathbb{N} be the set of positive integers. Recall that for any (non-empty) index \underline{s}=(s_{1},\ldots,s_{r})\in\mathbb{N}^{r} with s_{1}>1 , the classical multiple zeta value is defined by the infinite series \zeta(\underline{s})\mathrel{\mathop{\mathchar 58\relax}}=\sum_{n_{1}>\cdots>n_{r}\geq 1}\frac{1}{n_{1}^{s_{1}}\cdots n_{r}^{s_{r}}}\in\mathbb{R}. The quantities \operatorname{wt}(\underline{s})\mathrel{\mathop{\mathchar 58\relax}}=\sum_{i=1}^{r}s_{i} and \operatorname{dep}(\underline{s})\mathrel{\mathop{\mathchar 58\relax}}=r are called the weight and depth of the presentation \zeta(\underline{s}) , respectively. Via the integral (resp. series) representations of the multiple zeta values, one may express the product of two multiple zeta values as \mathbb{Q} -linear combinations of multiple zeta values of the same weight, called the shuffle (resp. stuffle) relations. In particular, this implies that the \mathbb{Q} -vector space spanned by all multiple zeta values admits a \mathbb{Q} -algebra structure. Furthermore, the above two product relations, together with some regularized process, lead to the so-called extended double shuffle relations of multiple zeta values. Conjecturally, these relations provide all \mathbb{Q} -linear relations among multiple zeta values [ IKZ06 ] . On the other hand, the notion of double Eisenstein series was introduced by Gangl, Kaneko, and Zagier [ GKZ06 ] , and the double zeta value in question appears as the constant term of the q -expansion of the corresponding double Eisenstein series. Bachmann [ Bac12 ] later generalized their idea and defined the multiple Eisenstein series as follows. Definition 1.1 ( [ GKZ06 ] ) . Let \tau be an element in the upper half-plane \mathbb{H}\mathrel{\mathop{\mathchar 58\relax}}=\{\tau\in\mathbb{C}\mid\operatorname{Im}\tau>0\} . For m,n\in\mathbb{Z}\tau+\mathbb{Z} , we define the notation \succ as follows: (1) m=a\tau+b\succ 0 if a>0 or (a,b)\in\{0\}\times\mathbb{N} . (2) m\succ n if m-n\succ 0 . Definition 1.2 ( [ GKZ06 ] for r\leq 2 , [ Bac12 ] for general r ) . For \underline{s}=(s_{1},\ldots,s_{r})\in\mathbb{N}^{r} with s_{1}\geq 3 and s_{2},\ldots,s_{r}\geq 2 , the multiple Eisenstein series is defined by G_{\underline{s}}(\tau)\mathrel{\mathop{\mathchar 58\relax}}=\sum_{\begin{subarray}{c}m_{1}\tau+n_{1},\ldots,m_{r}\tau+n_{r}\in\mathbb{Z}\tau+\mathbb{Z}\\ m_{1}\tau+n_{1}\succ\cdots\succ m_{r}\tau+n_{r}\succ 0\end{subarray}}\frac{1}{(m_{1}\tau+n_{1})^{s_{1}}\cdots(m_{r}\tau+n_{r})^{s_{r}}}. For \underline{s}=(s_{1},\ldots,s_{r}) with s_{1},\ldots,s_{r}\geq 2 and \tau\in\mathbb{H} , we consider the multitangent function [ Bou14 ] defined by \Psi_{\underline{s}}(\tau)\mathrel{\mathop{\mathchar 58\relax}}=\sum_{\begin{subarray}{c}n_{1},\ldots,n_{r}\in\mathbb{Z}\\ n_{1}>\cdots>n_{r}\end{subarray}}\frac{1}{(\tau+n_{1})^{s_{1}}\cdots(\tau+n_{r})^{s_{r}}}. By splitting up the summation in Definition 1.2 , one has the following expansion in terms of multitangent functions. Theorem 1.3 . For \underline{s}=(s_{1},\ldots,s_{r})\in\mathbb{N}^{r} with s_{1}\geq 3 and s_{2},\ldots,s_{r}\geq 2 , we have G_{\underline{s}}(\tau)=\zeta(\underline{s})+\sum_{j=1}^{r}\zeta(s_{j+1},\ldots,s_{r})g(s_{1},\ldots,s_{j}) where g is given by g(s_{1},\ldots,s_{r})\mathrel{\mathop{\mathchar 58\relax}}=\sum_{\begin{subarray}{c}1\leq j\leq r\\ 0=r_{0}<r_{1}<\cdots<r_{j-1}<r_{j}=r\\ m_{1}>\cdots>m_{j}>0\end{subarray}}\prod_{{}_{i=1}}^{j}\Psi_{s_{r_{i-1}+1},\ldots,s_{r_{i}}}(m_{i}\tau). Bachmann [ BB23 ] (see also [ Bac20 ] ) further expanded these multitangent functions and obtained the q -expansion of the multiple Eisenstein series G_{\underline{s}}(\tau) with the corresponding multiple zeta value \zeta(\underline{s}) appearing as the constant term, generalizing the result of Gangl-Kaneko-Zagier [ GKZ06 ] . The aforementioned q -expansion result leads to the questions of “lifting” relations from multiple zeta values to multiple Eisenstein series. However, some shuffle/stuffle relations cannot be lifted directly since the corresponding multiple Eisenstein series may not be defined due to the convergence issues. For example, in the shuffle relation \zeta(3)\zeta(3)=12\zeta(5,1)+6\zeta(4,2)+2\zeta(3,3) , the Eisenstein series is not defined for \underline{s}=(5,1) . This obstruction was overcome in [ BT18 ] and [ Bac19 ] by introducing the notions of shuffle and stuffle regularized multiple Eisenstein series , respectively. Their results can be described as the following. Theorem 1.4 ( [ BT18 ] for shuffle relations, [ Bac19 ] for stuffle relations) . The shuffle (resp. stuffle) regularized multiple Eisenstein series satisfy the same shuffle (resp. stuffle) relations as the corresponding multiple zeta values do. 1.2. Multiple Zeta Values in Positive Characteristic In this paper, we consider an analog of Theorem 1.4 in positive characteristic. Let p be a prime number and q be a power of p . Let A\mathrel{\mathop{\mathchar 58\relax}}=\mathbb{F}_{q}[\theta] be the polynomial ring in the variable \theta over a finite field \mathbb{F}_{q} of q elements (with the prime field \mathbb{F}_{p} ), and K\mathrel{\mathop{\mathchar 58\relax}}=\mathbb{F}_{q}(\theta) be its field of fractions. We let K_{\infty} be the completion of K with respect to the absolute value |\cdot| at the infinite place normalized so that |\theta|=q , and \mathbb{C}_{\infty} be the completion of a fixed algebraic closure of K_{\infty} . For a positive integer d , we let A_{+},A_{+,d},A_{<d} be the subsets of A consisting of all monic polynomials, monic polynomials of degree d , and polynomials of degree less than d , respectively. Moreover, we define A_{+,0}=\{1\} . In [ Tha04 ] , Thakur introduced the multiple zeta values over A , which generalized the notions of Carlitz single zeta values [ Car35 ] as follows. For any d\in\mathbb{Z}_{\geq 0} and (non-empty) index \underline{s}=(s_{1},\ldots,s_{r})\in\mathbb{N}^{r} , we define the d -th power sum (1) S_{d}(\underline{s})\mathrel{\mathop{\mathchar 58\relax}}=\sum_{\begin{subarray}{c}a_{1},\ldots,a_{r}\in A_{+}\\ d=\deg a_{1}>\cdots>\deg a_{r}\geq 0\end{subarray}}\frac{1}{a_{1}^{s_{1}}\cdots a_{r}^{s_{r}}}\in K and Thakur’s multiple zeta value \zeta_{A}(\underline{s})\mathrel{\mathop{\mathchar 58\relax}}=\sum_{d=0}^{\infty}S_{d}(\underline{s})=\sum_{\begin{subarray}{c}a_{1},\ldots,a_{r}\in A_{+}\\ \deg a_{1}>\cdots>\deg a_{r}\geq 0\end{subarray}}\frac{1}{a_{1}^{s_{1}}\cdots a_{r}^{s_{r}}}\in K_{\infty}. Similar to the classical setting, the quantities \operatorname{wt}(\underline{s})\mathrel{\mathop{\mathchar 58\relax}}=\sum_{i=1}^{r}s_{i} and \operatorname{dep}(\underline{s})\mathrel{\mathop{\mathchar 58\relax}}=r are called the weight and depth of the presentation \zeta_{A}(\underline{s}) , respectively. It is known by [ Tha09 ] that all these multiple zeta values are non-vanishing. Moreover, Thakur [ Tha10 ] showed the existence that the product of two multiple zeta values can be expressed as an \mathbb{F}_{p} -linear combination of multiple zeta values of the same weight, called the q -shuffle relations for multiple zeta values. In particular, this implies that the \mathbb{F}_{p} -vector space spanned by all multiple zeta values admits an \mathbb{F}_{p} -algebra structure. In contrast to the classical situation mentioned in Section 1.1 , Thakur [ Tha09 ] conjectured that the multiple zeta values over A admit only one “shuffle relation” and gave some heuristic reasons and observations based on the “motivic interpretations” of multiple zeta values [ AT09 ] . Regarding the existence of the q -shuffle product mentioned above, the first concrete example was due to Chen [ Che15 ] , who proved the following explicit formula: \zeta_{A}(r)\zeta_{A}(s)=\zeta_{A}(r,s)+\zeta_{A}(s,r)+\zeta_{A}(r+s)+\sum_{\begin{subarray}{c}i+j=r+s\\ q-1\mid j\end{subarray}}\Delta^{i,j}_{r,s}\zeta_{A}(i,j), where \Delta^{i,j}_{r,s}\mathrel{\mathop{\mathchar 58\relax}}=(-1)^{r-1}\binom{j-1}{r-1}+(-1)^{s-1}\binom{j-1}{s-1} . Later on, based on Chen’s formula, Yamamoto gave the following (conjectural) explicit q -shuffle relations for the multiple zeta values, which were verified in Shi’s PhD thesis [ Shi18 ] . Definition 1.5 . Let \mathcal{S} be the free monoid generated by the set \{x_{n}\mid n\in\mathbb{N}\} and let \mathcal{R} be the \mathbb{F}_{p} -vector space generated by \mathcal{S} . For \underline{s}=(s_{1},\ldots,s_{r})\in\mathbb{N}^{r} we define x_{\underline{s}}\mathrel{\mathop{\mathchar 58\relax}}=x_{s_{1}}\cdots x_{s_{r}} , and for \underline{s}=\varnothing we define x_{\varnothing}\mathrel{\mathop{\mathchar 58\relax}}=1 . An element x_{\underline{s}} is called a word of depth r . We define the q -shuffle product \ast on \mathcal{R} inductively on the sum of depths as follows: (1) For the empty word x_{\varnothing}=1 and any \mathfrak{a}\in\mathcal{R} , define 1*\mathfrak{a}=\mathfrak{a}*1=\mathfrak{a}. (2) For \underline{r}=(r_{1},\ldots,r_{m})\in\mathbb{N}^{m} and \underline{s}=(s_{1},\ldots,s_{n})\in\mathbb{N}^{r} , define x_{\underline{r}}*x_{\underline{s}}=x_{r_{1}}(x_{\underline{r}^{-}}*x_{\underline{s}})+x_{s_{1}}(x_{\underline{r}}*x_{\underline{s}^{-}})+x_{r_{1}+s_{1}}(x_{\underline{r}^{-}}*x_{\underline{s}^{-}})\\ +\sum_{\begin{subarray}{c}i+j=r_{1}+s_{1}\\ q-1\mid j\end{subarray}}\Delta^{i,j}_{r_{1},s_{1}}x_{i}((x_{\underline{r}^{-}}*x_{\underline{s}^{-}})*x_{j}) where (a) \Delta^{i,j}_{r,s}\mathrel{\mathop{\mathchar 58\relax}}=(-1)^{r-1}\binom{j-1}{r-1}+(-1)^{s-1}\binom{j-1}{s-1} . (b) x\sum_{i}f_{i}x_{i}\mathrel{\mathop{\mathchar 58\relax}}=\sum_{i}f_{i}(xx_{i}) for f_{i}\in\mathbb{F}_{p} and x,x_{i}\in\mathcal{S} . (c) x_{\underline{r}^{-}}\mathrel{\mathop{\mathchar 58\relax}}=x_{r_{2}}\cdots

Read the complete text