Alphabeta Math
LemmaStatement: AI-adaptedProof: AI-generatedPipeline-generatedprecheck passjudge pass (gpt-6-luna)verified 2026-11-04 (gpt-6.1-sol)
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Hurwitz Zeta Values for RNS Normal Ordering

Statement

For the Hurwitz zeta function ζ(s,a), continued meromorphically from ∑n=0∞(n+a)−s with a>0, and B2(a)=a2−a+16, one has ζ(−1,a)=−12B2(a). In particular, ζ(−1,1)=−1/12 and ζ(−1,1/2)=1/24.

Facts & Assumptions

Given: Fix real a>0. For Re⁡s>1, define ζ(s,a)=∑n=0∞(n+a)−s and use its meromorphic continuation. Let B~4(t)=B4(t−⌊t⌋) be the bounded periodic fourth Bernoulli function; Euler–Maclaurin with two Bernoulli corrections continues the tail integral to a neighborhood of s=−1.

[A1]

B2(a)=a2−a+1/6 and B4(x)=x4−2x3+x2−1/30; the Hurwitz zeta uses the positive-real branch of (n+a)−s.

Proof

1.1givenA1algebra

Fix an integer N≥1, put x=N+a, and take f(t)=t−s for Re⁡s>1 and t≥x. Then f is C4, its derivatives through order three tend to zero at infinity, and f(4) is integrable. Four integrations by parts on each interval [x+n,x+n+1] against the periodic Bernoulli polynomials, summed over n≥0, give the Euler–Maclaurin tail identity ∑n=0∞f(x+n)=∫x∞f(t) dt+12f(x)−B22!f′(x)−B44!f′′′(x)−14!∫x∞B~4(t−x)f(4)(t) dt, where B2=1/6, B4=−1/30, and B~4(t)=B4(t−⌊t⌋). The endpoint terms telescope because the periodic Bernoulli polynomials agree at consecutive integer endpoints; the last integral is the remaining integration-by-parts term. Here ∫x∞t−s dt=x1−s/(s−1), and substitution of the derivatives gives ζ(s,a)=∑n=0N−1(n+a)−s+x1−ss−1+12x−s+s12x−s−1−s(s+1)(s+2)720x−s−3+R2(s,x), with R2(s,x)=−s(s+1)(s+2)(s+3)24∫x∞B~4(t−x)t−s−4 dt. The integral converges locally uniformly for Re⁡s>−3, so this expression continues the defining series there (apart from its pole at s=1). At s=−1, both the third-derivative correction and R2 vanish because they contain a factor s+1.

2.1givenA1step 1.1algebra

At s=−1, the finite sum in step 1.1 is ∑n=0N−1(n+a)=N(N−1)/2+Na, the integral term is −x2/2, the endpoint term is x/2, and the B2 correction is −1/12. The B4 correction and remainder vanish because they contain a factor s+1. Therefore ζ(−1,a)=N(N−1)2+Na−(N+a)22+N+a2−112=−12(a2−a+16)=−12B2(a). The right side is independent of N, as required for the continuation value.

3.1givenA1step 2.1algebra∎

Substitution gives B2(1)=1/6, so ζ(−1,1)=−1/12. Also B2(1/2)=1/4−1/2+1/6=−1/12, so ζ(−1,1/2)=1/24. These are exactly the integer and half-integer vacuum sums used in the oscillator calculation.

Remarks

The NIST formula 25.11.14 independently records the general negative-integer value. The proof above uses the Euler–Maclaurin continuation directly, so the RNS intercept calculation does not take its own target value as a premise.

Dependency tree · 0 levels

Nothing. This result depends on no other item in the library.

Sources