Sums of Powers of Sine and Generalized Bernoulli Polynomials

Leon D. Fairbanks

arxiv: 2502.15966 · v5 · submitted 2025-02-21 · 🧮 math.GM

Sums of Powers of Sine and Generalized Bernoulli Polynomials

Leon D. Fairbanks This is my paper

Pith reviewed 2026-05-23 02:03 UTC · model grok-4.3

classification 🧮 math.GM

keywords sums of powers of sinegeneralized Bernoulli polynomialsEuler polynomialsRiemann zeta functionintegral representationdyadic anglestrigonometric sums

0 comments

The pith

Sums of reciprocal powers of sines at dyadic angles equal closed forms in generalized Bernoulli and Euler polynomials, which also produce an integral representation of the Riemann zeta function.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper derives explicit formulas expressing the finite sum from j equals 1 to 2 to the n minus 2 of one over sine to the s of (2j minus 1) pi over 2 to the n as combinations of generalized Bernoulli and Euler polynomials. It then converts one of those formulas into an integral representation for the Riemann zeta function. A sympathetic reader would care because these particular trigonometric sums arise in Fourier analysis and number theory, and closed polynomial expressions supply a new route to evaluate or relate them. The work focuses on the dyadic angles that make the sums amenable to polynomial identities rather than infinite series or other methods.

Core claim

The central claim is that the sum ∑_{j=1}^{2^{n-2}} 1/sin^s((2j-1)π/2^n) admits closed-form expressions in the generalized Bernoulli and Euler polynomials for positive integer s, and that substituting one such expression produces an integral representation of the Riemann zeta function.

What carries the argument

Generalized Bernoulli and Euler polynomials, which serve as the closed-form containers for the trigonometric sums at the specified dyadic angles.

If this is right

The sums equal explicit polynomial combinations rather than requiring infinite products or series.
One of the polynomial expressions converts directly into an integral equal to the Riemann zeta function.
The same polynomial machinery applies uniformly for each fixed power s across varying n.
The formulas relate finite trigonometric sums to objects already studied in number theory.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the polynomial expressions hold, they could be differentiated or integrated termwise to obtain identities for related sums or products.
The integral representation of zeta might be combined with other known integrals to produce new functional equations or bounds.
Similar closed forms could exist for sums at other rational multiples of pi if the same polynomial families are adjusted by a change of variables.

Load-bearing premise

The specific finite sums over these dyadic angles admit closed forms in the generalized Bernoulli and Euler polynomials.

What would settle it

Direct numerical evaluation of the sum for small fixed n and s compared against the corresponding polynomial expression; disagreement for any such pair falsifies the claimed formulas.

read the original abstract

We produce formulas for $$\sum_{j=1}^{2^{n-2}}\frac{1}{\sin^s\left(\frac{(2j-1)\pi}{2^n}\right)}$$ in terms of Generalized Bernoulli and Euler polynomials and use one of the formulas to produce a nice integral representation of the Riemann zeta function.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The abstract asserts closed forms for those exact dyadic sine sums in generalized Bernoulli and Euler polynomials plus a zeta integral, but supplies zero derivations or checks.

read the letter

The paper's central assertion is that the finite sums over 1/sin^s at the angles (2j-1)pi/2^n for j from 1 to 2^{n-2} can be written in closed form using generalized Bernoulli and Euler polynomials, and that one such expression yields an integral representation for zeta. That is the one or two things to know up front. The connection between power sums of sines at roots-of-unity angles and Bernoulli-type polynomials is a standard technique, so the paper is applying existing machinery to this narrow family of sums. If the explicit formulas turn out to be correct and reasonably compact, that would be a modest but concrete addition to the catalog of such identities. The zeta integral is presented as a byproduct, which could be clean if the steps avoid hidden assumptions about the range of s or the limiting process. The main limitation is that only the abstract is available. No actual formulas, no derivation outline, no error bounds, and no literature comparison appear, so it is impossible to verify whether the claimed expressions are new, correct, or free of circularity. The sums themselves are tightly scoped to these dyadic angles, which limits how far any result can travel even if it holds. This kind of work is mainly of interest to specialists who already track explicit identities involving Bernoulli polynomials and trigonometric sums. A reader hunting for general methods or broad applications to zeta will not find enough here to engage. I would not send the paper to peer review until the full derivations are supplied and can be examined; at that point a referee could check novelty and correctness directly.

Referee Report

2 major / 0 minor

Summary. The manuscript claims to produce closed-form expressions for the finite sums ∑_{j=1}^{2^{n-2}} [1/sin((2j-1)π/2^n)]^s in terms of generalized Bernoulli and Euler polynomials (for positive integers n and suitable s), and to derive from one such expression an integral representation of the Riemann zeta function.

Significance. If the claimed closed forms are correct and independently verifiable, the work would connect standard techniques for power sums over roots of unity to the generalized Bernoulli/Euler families, potentially yielding new evaluations or zeta representations. The integral representation for zeta would be of interest if it is non-trivial and rigorously derived. However, the manuscript supplies neither explicit formulas nor derivations, so the significance cannot be assessed from the provided content.

major comments (2)

No explicit formulas, derivations, error bounds, or worked examples are supplied for the claimed sums in terms of generalized Bernoulli or Euler polynomials, preventing verification of the central claim that these particular dyadic-angle sums admit such closed forms.
The manuscript contains no derivation or justification for the integral representation of the Riemann zeta function obtained from one of the formulas, so it is impossible to check whether the representation is valid, novel, or correctly obtained from the trigonometric sum.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their report and for identifying the gaps in explicit content. We will revise the manuscript to address both major comments by adding the requested formulas, derivations, and justifications.

read point-by-point responses

Referee: No explicit formulas, derivations, error bounds, or worked examples are supplied for the claimed sums in terms of generalized Bernoulli or Euler polynomials, preventing verification of the central claim that these particular dyadic-angle sums admit such closed forms.

Authors: We agree that the current manuscript does not supply the explicit closed-form expressions, step-by-step derivations, error bounds, or worked examples. In the revised version we will insert the explicit formulas expressing the sums in terms of generalized Bernoulli and Euler polynomials, provide the derivations from the relevant generating functions or recurrence relations, include error bounds where the expressions are approximate, and add numerical checks for small values of n and s to permit direct verification. revision: yes
Referee: The manuscript contains no derivation or justification for the integral representation of the Riemann zeta function obtained from one of the formulas, so it is impossible to check whether the representation is valid, novel, or correctly obtained from the trigonometric sum.

Authors: We acknowledge the absence of any derivation or justification for the integral representation of zeta(s). The revised manuscript will contain a dedicated section that starts from the closed-form sum, applies the appropriate limiting or integral transform procedure, and arrives at the claimed integral representation, together with a brief discussion of its validity, the conditions on s, and its relation to known integral representations of zeta. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The abstract states that formulas are produced for the indicated finite sums in terms of generalized Bernoulli and Euler polynomials, with one used for a zeta integral representation. No equations, self-citations, fitted parameters, or ansatzes are visible that reduce the claimed results to their inputs by definition or construction. The connection between power sums over dyadic sine arguments and these polynomials is a standard technique relying on external properties of roots of unity and generating functions, not on any internal loop or self-referential definition within the paper. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review; ledger populated from claims in the abstract. No explicit free parameters, ad-hoc axioms, or invented entities are stated.

axioms (1)

domain assumption Generalized Bernoulli and Euler polynomials satisfy the identities needed to express the indicated trigonometric sums
Invoked in the first sentence of the abstract to produce the formulas.

pith-pipeline@v0.9.0 · 5561 in / 1061 out tokens · 29751 ms · 2026-05-23T02:03:26.908449+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Theorem 4.1: S(2m+1,n) expressed via Euler polynomials E_{2m-2k}(j/2^{n-1}) and generalized Bernoulli B^{(2m)}_{2k}(m)
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

zeta integral (4) obtained by taking n→∞ limit of the odd-power formula

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

6 extracted references · 6 canonical work pages · 1 internal anchor

[1]

arXiv:2502.15966v4 [math.GM] 1 May 2025 Sums of Powers of Sine and Generalized Bernoulli Polynomials Leon Fairbanks May 5, 2025 Abstract We produce formulas for 2n−2 ∑ j=1 1 sins ( (2 j−1)π 2n ) in terms of Generalized Bernoulli and Euler polynomials and use one of the formulas to produce a nice integral representation of the Ri emann zeta function. Conte...

work page internal anchor Pith review Pith/arXiv arXiv 2025
[2]

∫ 1/2 0 E2m(x) csc(πx)dx ■ 5 Sums Of Negative Even Powers As previously noted, if we let S(s, n) = 2n−2 ∑ j=1 1 sins ( (2 j−1)π 2n ) 38 Then S(2, n) = 1 2 22n−2 S(4, n) = 1 6 ( 24n−4 + 2(22n−2) ) S(6, n) = 1 30 (2(26n−6) +5(24n−4) +8(22n−2)) S(8, n) = 1 630 ( 17(28n−8) +56(26n−6) +98(24n−4) +144(22n−2) ) We can see a pattern if we write these as S(2, n) =...

work page 1916
[3]

See Prop

π 2 j(2(k − j))! B(2k) 2(k− j)(k)ζ (2 j) (99) 39 Equivalently, S(2k, n) = (−1)k+1 22k−1 (2k)! k ∑ j=1 22 jn 22 j − 1 22 j ( 2k 2 j ) B(2k) 2(k− j)(k)B(2 j) (100) (101) Proof. See Prop. 5.6 and the comments preceding the proposition. ■ Here’s what this looks like. S(2, n) = 3 π 2 ζ (2)22n−2 S(4, n) = 15 π 4 ζ (4)24n−4 + 2 π 2 ζ (2)22n−2 S(6, n) = 63 π 6 ζ ...

work page 2046
[4]

Fairbanks, “Notes On An Approach to Apery’s Constant ’, arXiv:2206.11256 [math.NT] 44 •

L. Fairbanks, “Notes On An Approach to Apery’s Constant ’, arXiv:2206.11256 [math.NT] 44 •

work page arXiv
[5]

(1972), ”Stirl ing Numbers of the First Kind, Stirling Numbers of the Second Kind”, in Abra mowitz, Milton; Stegun, Irene A

Goldberg, K.; Newman, M; Haynsworth, E. (1972), ”Stirl ing Numbers of the First Kind, Stirling Numbers of the Second Kind”, in Abra mowitz, Milton; Stegun, Irene A. (eds.), Handbook of Mathematical Function s with Formulas, Graphs, and Mathematical Tables, 10th printing, New Y ork: Dover, pp. 824–825 •

work page 1972
[6]

Srivastava, A

H.M. Srivastava, A. Pinter.,”Remarks on Some Relation ships Between the Bernoulli and Euler Polynomials”, Applied Mathematics Let ters 17 (2004) 375-380 45

work page 2004

[1] [1]

arXiv:2502.15966v4 [math.GM] 1 May 2025 Sums of Powers of Sine and Generalized Bernoulli Polynomials Leon Fairbanks May 5, 2025 Abstract We produce formulas for 2n−2 ∑ j=1 1 sins ( (2 j−1)π 2n ) in terms of Generalized Bernoulli and Euler polynomials and use one of the formulas to produce a nice integral representation of the Ri emann zeta function. Conte...

work page internal anchor Pith review Pith/arXiv arXiv 2025

[2] [2]

∫ 1/2 0 E2m(x) csc(πx)dx ■ 5 Sums Of Negative Even Powers As previously noted, if we let S(s, n) = 2n−2 ∑ j=1 1 sins ( (2 j−1)π 2n ) 38 Then S(2, n) = 1 2 22n−2 S(4, n) = 1 6 ( 24n−4 + 2(22n−2) ) S(6, n) = 1 30 (2(26n−6) +5(24n−4) +8(22n−2)) S(8, n) = 1 630 ( 17(28n−8) +56(26n−6) +98(24n−4) +144(22n−2) ) We can see a pattern if we write these as S(2, n) =...

work page 1916

[3] [3]

See Prop

π 2 j(2(k − j))! B(2k) 2(k− j)(k)ζ (2 j) (99) 39 Equivalently, S(2k, n) = (−1)k+1 22k−1 (2k)! k ∑ j=1 22 jn 22 j − 1 22 j ( 2k 2 j ) B(2k) 2(k− j)(k)B(2 j) (100) (101) Proof. See Prop. 5.6 and the comments preceding the proposition. ■ Here’s what this looks like. S(2, n) = 3 π 2 ζ (2)22n−2 S(4, n) = 15 π 4 ζ (4)24n−4 + 2 π 2 ζ (2)22n−2 S(6, n) = 63 π 6 ζ ...

work page 2046

[4] [4]

Fairbanks, “Notes On An Approach to Apery’s Constant ’, arXiv:2206.11256 [math.NT] 44 •

L. Fairbanks, “Notes On An Approach to Apery’s Constant ’, arXiv:2206.11256 [math.NT] 44 •

work page arXiv

[5] [5]

(1972), ”Stirl ing Numbers of the First Kind, Stirling Numbers of the Second Kind”, in Abra mowitz, Milton; Stegun, Irene A

Goldberg, K.; Newman, M; Haynsworth, E. (1972), ”Stirl ing Numbers of the First Kind, Stirling Numbers of the Second Kind”, in Abra mowitz, Milton; Stegun, Irene A. (eds.), Handbook of Mathematical Function s with Formulas, Graphs, and Mathematical Tables, 10th printing, New Y ork: Dover, pp. 824–825 •

work page 1972

[6] [6]

Srivastava, A

H.M. Srivastava, A. Pinter.,”Remarks on Some Relation ships Between the Bernoulli and Euler Polynomials”, Applied Mathematics Let ters 17 (2004) 375-380 45

work page 2004