cert-machine · one of six · re-verified from the manuscript

Erdős Problem #1038

The infimum is exactly D = 1.834430475762661711090753635125…, in a July 2026 manuscript of Darvas, Peng and Tao. Among monic polynomials with all roots in [−1,1], what is the infimum of the measure of the set where |f| < 1? Open since Erdős–Herzog–Piranian, 1958.

tl;dr
  • The finding. CONFIRMED — the computational fragment. What a machine can decide about this claim, it decided, and it held. What it cannot decide is stated below rather than left to the reader to notice.
  • The mechanism. Written against the manuscript. The author's code was never executed. Verification that existed at source: Author-checked manuscript shipping its own three Arb certificates and two SymPy verifiers; the official Erdős-problems record was still open.
  • Check it. node legacy/research/challenges/lane/laneb-lemniscate/verify.js — 0 ms at this build, 22 named checks and 4 mutation controls, every control rejected.
verdict
CONFIRMED
the computational fragment
named checks
22
each one printed by the verifier at this build and listed on this page
the verifier's own total
26
its total folds the 4 mutation controls in; ours does not
mutation controls
4
deliberate corruptions of the claim — every one rejected, or this page would not build
runtime
0 ms
plain Node, no packages, no author code
analytic core
not audited
true of all six claims in this set, and the reason the flagship exists
§1 · at source

What was claimed, by whom, and what checking already existed

fieldas recorded at source
the claimThe infimum is exactly D = 1.834430475762661711090753635125…, in a July 2026 manuscript of Darvas, Peng and Tao.
the problemAmong monic polynomials with all roots in [−1,1], what is the infimum of the measure of the set where |f| < 1? Open since Erdős–Herzog–Piranian, 1958.
credited AI systemGPT-5.5 Pro (initial main argument in a solver–verifier framework, then revised by the human authors)
verification at sourceAuthor-checked manuscript shipping its own three Arb certificates and two SymPy verifiers; the official Erdős-problems record was still open.
our verdictCONFIRMED the computational fragment
§2 · what we certified

The ledger, as the verifier printed it at this build

The extremal triple exists, is unique inside a certified box, lies in the admissible region, and every printed decimal of the extremal constants is correct — including all 30 claimed decimals of D.

This verifier prints one flat list rather than stages. All 22 named rows follow, in the order it emitted them.

what the verifier proved
1(A.4) AD d/du F_R overlaps the printed closed form [AD -5.2402332, closed -5.2402332]
2(A.5) AD d/dv F_L overlaps the printed closed form [AD 84.393443, closed 84.393443]
3(A.6) AD d/du X_q(u) overlaps the printed closed form [AD -8.6686021, closed -8.6686021]
4(A.12) 0 < p(q) < q < 1 uniformly on I_* [p in 0.824521729252520379285 .. 0.824521729252520379285]
5(A.13) F_R(q, u0-1e-30) > 0 > F_R(q, u0+1e-30) on I_* (mean-value form) [F_R faces 0.00000000000000000000000000000405 .. 0.00000000000000000000000000000642 | -0.00000000000000000000000000000642 .. -0.00000000000000000000000000000405]
6(A.14) F_L(q, v0-1e-30) < 0 < F_L(q, v0+1e-30) on I_* (mean-value form) [F_L faces -0.00000000000000000000000000009038 .. -0.00000000000000000000000000007839 | 0.00000000000000000000000000007839 .. 0.00000000000000000000000000009038]
7Krawczyk (eqcert) proves existence + uniqueness of the triple [rounds 1, maxRad 3.84e-14]
8B_* (A.9) is contained in the Krawczyk uniqueness box
9high-precision Krawczyk image lands in the interior (existence re-proved)
10refined enclosure width < 1e-40 in each coordinate [widths(quanta 1e-48): 1636, 3183, 1124]
11q* inside I_* = q0 ± 1e-31 (A.11) — stronger than the B_* face
12(q*,u*,v*) inside B_* = centres ± 1e-30 (A.9) => Lemma A.1 confirmed
13claimed digits D (p.2, 30 dec) [rounded]
14claimed digits alpha (p.5) [truncation]
15claimed digits q* (p.24) [truncation]
16claimed digits u* (p.24) [truncation]
17claimed digits v* (p.24) [truncation]
18claimed digits p* (p.24) [truncation]
19claimed digits alpha (p.24) [truncation]
20claimed digits r* (p.24) [truncation]
21claimed digits z* (p.24) [truncation]
22claimed digits D (p.24) [truncation]
§3 · the falsifiers

Proof that this verifier can fail

A verifier that cannot reject a false claim is not evidence, it is decoration. So each one carries mutation controls: the claim is deliberately corrupted and the verifier must refuse it. All 4 were rejected in the run that built this page, and a control that stopped firing would refuse the page instead.

the corruption, and how it was caught
1M1 rejected: D last digit tampered (…125 -> …126) fails the digit check
2M2 rejected: K with flipped sign has NO zero at the certified triple [F3_mut = 1.307440418726 .. 1.307440418726]
3M3 rejected: box shifted by 1e-10 in q contains no zero of F_R [F_R over shifted box = 0.00000000118560 .. 0.00000000118560]
4M4 rejected: p(q) constant 2 -> 2.000000001 breaks F_R at the triple [F_R_mut = -0.00000000047753 .. -0.00000000047753]
§4 · the boundary

What this audit did NOT reach

not audited

The proof that this triple is extremal. We certified the number the manuscript names; we did not certify that no polynomial does better.

This is the part worth reading twice. The verdict at the top of this page is true inside its scope line and false outside it, and the sentence above is where the scope line comes from. Across all six claims in this set the same split appears: the computational fragment certifies and the analytic core does not — the flagship page measures exactly that.

§5 · re-run it

One command, no dependencies

git clone https://github.com/carlostoledo1891/cert-machine
cd cert-machine
node legacy/research/challenges/lane/laneb-lemniscate/verify.js

Plain Node, no packages. It prints the ledger above, runs its own falsifiers, and exits non-zero if anything fails. node instruments/laneaudit/audit.js runs all six.