OpenFrontiers
← Back to the index
OP-12073OpenAnalysis

Bohr radius for the bidisc (C59C_{59})

Bohr radius for the bidisc: determine C59C_{59} or improve its bounds; the repository records 0.3006 (lower) and 0.302825279492 (upper).

Editorial MSC2020 classification. Original article codes and source evidence are recorded in the literature directory.

Licensed source statementAs of Repository table at commit 2c1968c (2026-09-26); open while the recorded lower and upper bounds differSource & evidence ↓

Problem statement

Let

Dd := {z=(z1,…,zd)∈Cd: ∣z1∣,…,∣zd∣<1}\mathbb{D}^d\ :=\ \{z=(z_1,\dots,z_d)\in\mathbb{C}^d:\ \lvert z_1\rvert,\dots,\lvert z_d\rvert<1\}

be the unit polydisc, and let the Schur class SdS_d be the set of analytic functions f:Dd→Df:\mathbb{D}^d\to\mathbb{D}. [Kne2025-def-polydisc] [Kne2025-def-Schur]

Writing the power series expansion f(z)=∑α∈N0dfαzαf(z)=\sum_{\alpha\in\mathbb{N}_0^d} f_\alpha z^\alpha, define the coefficient-wise ℓ1\ell^1 norm ∥f∥1:=∑α∣fα∣\lVert f\rVert_1:=\sum_\alpha \lvert f_\alpha\rvert and the dilation fr(z):=f(rz)f_r(z):=f(rz). [Kne2025-def-l1] [Kne2025-def-fr]

The Bohr radius KdK_d is defined by

Kd := sup⁡{r>0: ∥fr∥1≤1 for all f∈Sd}.K_d\ :=\ \sup\Bigl\{r>0:\ \lVert f_r\rVert_1\le 1\ \text{for all } f\in S_d\Bigr\}.

[Kne2025-def-Kd]

Equivalently, KdK_d is the largest number such that for every power series ∑αcαzα\sum_\alpha c_\alpha z^\alpha with ∣∑αcαzα∣<1\bigl\lvert\sum_\alpha c_\alpha z^\alpha\bigr\rvert<1 on Dd\mathbb{D}^d, one has ∑α∣cαzα∣<1\sum_\alpha \lvert c_\alpha z^\alpha\rvert<1 whenever max⁡1≤j≤d∣zj∣<Kd\max_{1\le j\le d}\lvert z_j\rvert<K_d. [BK1997-def-Kn]

We define

C59 := K2,C_{59}\ :=\ K_2,

the Bohr radius for the bidisc D2\mathbb{D}^2.

Bohr’s one-variable theorem gives K1=1/3K_1=1/3, and in particular implies K2≤1/3K_2\le 1/3. [BK1997-Bohr-1d] [BK1997-ub-1-3]

The exact value of KdK_d is unknown for every d>1d>1; in particular, the exact value of K2K_2 is open. [BK1997-open]

The best established range currently is

0.3006 ≤ K2 < 0.302825279492.0.3006\ \le\ K_2\ <\ 0.302825279492.

[Kne2025-lb-K2-0-3006] [P2026-ub-K2-0-302825279492]

Lean 4

Not yet attached

No Lean statement or proof is attached yet. Have a formalization? Share its source to add it to this problem.

Background

  • Asymptotics in high dimension. The Bohr radius satisfies Kd≍(log⁡d)/dK_d\asymp \sqrt{(\log d)/d} as d→∞d\to\infty (up to absolute constants). [Kne2025-asymp-Kd]

References (from the repository page)

  • [BK1997] Boas, Harold P.; Khavinson, Dmitry. Bohr’s power series theorem in several variables. Proceedings of the American Mathematical Society 125 (1997), no. 10, 2975–2979. DOI: https://doi.org/10.1090/S0002-9939-97-04270-6. arXiv PDF: https://arxiv.org/pdf/math/9606203. Google Scholar

  • [BK1997-Bohr-1d] loc: arXiv v1 PDF p.1, Theorem 1 quote: “Then ∑k=0∞∣ckzk∣<1\sum_{k=0}^\infty \lvert c_k z^k\rvert<1 when ∣z∣<1/3\lvert z\rvert<1/3. Moreover, the radius 1/31/3 is the best possible.”

  • [BK1997-def-Kn] loc: arXiv v1 PDF p.1, definition paragraph for KnK_n quote: “Let KnK_n denote the n-dimensional Bohr radius: the largest number such that if ∑αcαzα\sum_\alpha c_\alpha z^\alpha converges in the unit polydisc {(z1,…,zn):max⁡1≤j≤n∣zj∣<1}\{(z_1,\ldots,z_n):\max_{1\le j\le n}\lvert z_j\rvert<1\}, and if ∣∑αcαzα∣<1\left\lvert\sum_\alpha c_\alpha z^\alpha\right\rvert<1 in the unit polydisc, then ∑α∣cαzα∣<1\sum_\alpha \lvert c_\alpha z^\alpha\rvert<1 when max⁡1≤j≤n∣zj∣<Kn\max_{1\le j\le n}\lvert z_j\rvert<K_n.”

  • [BK1997-ub-1-3] loc: arXiv v1 PDF p.2, paragraph after definition of KnK_n quote: “It is evident from Bohr’s one-dimensional result that Kn≤1/3K_n\le 1/3 for every nn.”

  • [BK1997-lb-1-3sqrt] loc: arXiv v1 PDF p.2, Proof of Theorem 2 quote: “This ball evidently contains the polydisc {z:max⁡1≤j≤n∣zj∣<1/(3n)}\{z:\max_{1\le j\le n}\lvert z_j\rvert<1/(3\sqrt{n})\}, whence Kn≥1/(3n)K_n\ge 1/(3\sqrt{n}).”

  • [BK1997-open] loc: arXiv v1 PDF p.2, Open question quote: “Open question. What is the exact value of the Bohr radius KnK_n when n>1n>1?”

  • [Kne2025] Knese, Greg. Three radii associated to Schur functions on the polydisk. Proceedings of the American Mathematical Society, Series B 12 (2025), no. 5, 48–63. DOI: https://doi.org/10.1090/bproc/262. arXiv PDF: https://arxiv.org/pdf/2410.21693. Google Scholar

  • [Kne2025-def-polydisc] loc: arXiv v3 PDF p.1, Introduction quote: “Dd={z=(z1,…,zd)∈Cd: ∣z1∣,…,∣zd∣<1}\mathbb{D}^d=\{z=(z_1,\dots,z_d)\in\mathbb{C}^d:\ \lvert z_1\rvert,\dots,\lvert z_d\rvert<1\}.”

  • [Kne2025-def-Schur] loc: arXiv v3 PDF p.1, Introduction quote: “The Schur class SdS_d of the polydisk Dd\mathbb{D}^d is the set of all analytic f:Dd→Df:\mathbb{D}^d\to\mathbb{D}.”

  • [Kne2025-def-l1] loc: arXiv v3 PDF p.1, equation (1.1) context in Introduction quote: “Define the coefficient-wise ℓ1\ell^1 norm ∥f∥1:=∑α∣fα∣\lVert f\rVert_1:=\sum_\alpha \lvert f_\alpha\rvert.”

  • [Kne2025-def-fr] loc: arXiv v3 PDF p.1, Introduction quote: “For r>0r>0 define fr(z):=f(rz)f_r(z):=f(rz).”

  • [Kne2025-def-Kd] loc: arXiv v3 PDF p.2, Introduction quote: “Define the Bohr radius KdK_d by Kd:=sup⁡{r>0: ∥fr∥1≤1 for all f∈Sd}K_d:=\sup\{r>0:\ \lVert f_r\rVert_1\le 1\ \text{for all } f\in S_d\}.”

  • [Kne2025-lb-K2-0-3006] loc: arXiv v3 PDF, Corollary 1.2 quote: “Corollary 1.2. K(A2)=K2≥0.3006K(\mathcal{A}_2)=K_2\ge 0.3006.”

  • [Kne2025-asymp-Kd] loc: arXiv v3 PDF p.2, Introduction quote: “After the culmination of deep work by many authors the precise asymptotic Kd∼log⁡d/dK_d\sim \sqrt{\log d/d} was established; see [18], [8].”

  • [BPWW2026] Baran, Radomił; Pikul, Piotr; Woerdeman, Hugo J.; Wojtylak, Michał. Contractive realization theory for the annulus and other intersections of disks on the Riemann sphere. Journal of Functional Analysis 290 (2026), no. 8, 111346. DOI: https://doi.org/10.1016/j.jfa.2026.111346. arXiv PDF: https://arxiv.org/pdf/2504.03236. Google Scholar

  • [BPWW2026-known-interval] loc: arXiv v1 PDF p.2, Introduction quote: “The constant K2K_2 is the 2-variate version of the Bohr constant, and is known to lie in the interval (0.3006,1/3)(0.3006,1/3). We are able to narrow the interval to (0.3006,0.3177)(0.3006,0.3177) in Theorem 6.4.”

  • [BPWW2026-ub-K2-0-3177] loc: arXiv v1 PDF p.18, Theorem 6.4 quote: “Theorem 6.4. K2<0.3177K_2<0.3177.”

  • [G2026] Griego, Sebastian. Rational-inner Fejer averaging certificate for the bidisc Bohr radius bound K2<0.3174541K_2<0.3174541, submitted to this repository (2026).

  • [G2026-ub-K2-0-3174541] loc: pull request certificate and exact verifier quote: “The exact integer verifier proves B3174541/10000000(p)>1+1/7307638490B_{3174541/10000000}(p)>1+1/7307638490 and the rational-inner Fejer averaging certificate proves ∣p∣≤1\lvert p\rvert\le 1 on the bidisc.”

  • [P2026] Patel, Shivam. Complex phases and a certified upper bound for the bidisc Bohr radius. Research note and reproducibility record, 26 August 2026. MathDB solution. Exact Python and Lean proof package, v1.0.0, Zenodo, 5 September 2026, DOI: 10.5281/zenodo.22341928; GitHub release; pinned source.

  • [P2026-ub-K2-0-302825279492] loc: the analytic and finite coefficient certificates in the package README; certificate/EndToEnd.lean, theorem Optim.BohrRadius.bohrRadius_lt_302825279492_div_10pow12 statement: The explicit Schur witness and exact finite coefficient comparison prove K2<302825279492/1012=0.302825279492K_2<302825279492/10^{12}=0.302825279492. The exact value of K2K_2 remains open.

Progress & source assessment

Repository table (commit 2c1968c, 26 September 2026): best known lower bound 0.3006; best known upper bound 0.302825279492. Values marked with an asterisk have minimal verification.

Known upper bounds

Bound Reference Comments
1/31/3 [BK1997] General upper bound Kn≤1/3K_n\le 1/3 (hence K2≤1/3K_2\le 1/3). [BK1997-ub-1-3]
0.31770.3177 [BPWW2026] Explicit construction giving K2<0.3177K_2<0.3177 (Theorem 6.4). [BPWW2026-ub-K2-0-3177]
0.31745410.3174541 [G2026] Degree-(250,250)(250,250) polynomial from a rational-inner Fejer averaging certificate. Exact integer verification gives B3174541/10000000(p)>1B_{3174541/10000000}(p)>1. [G2026-ub-K2-0-3174541]
0.3028252794920.302825279492 [P2026] Set U=(1+z)(1−w)U=(1+z)(1-w), V=1+zwV=1+zw, P=LU+iTVP=LU+iTV, Q=LV+iTUQ=LV+iTU, and f=(SQ−P)/(SQ+P)f=(SQ-P)/(SQ+P), with L=2500000000L=2500000000, T=3067398171T=3067398171, and S=1015S=10^{15}. Two independent exact computations of the 841 coefficients with 0≤j,k≤N=280\le j,k\le N=28 give a finite majorant greater than 1+10−261+10^{-26} at r=302825279492/1012r=302825279492/10^{12}. Proof package v1.0.0, DOI: 10.5281/zenodo.22341928; archived bidisc-bohr-certificate-v1.0.0.zip SHA-256: afc77b42cdd9de9b82e3d4c6a973dc32bf82b2c19960df58507b9a5cf333928b. [P2026-ub-K2-0-302825279492]

Known lower bounds

Bound Reference Comments
1/(32)1/(3\sqrt{2}) [BK1997] Special case of Kn≥1/(3n)K_n\ge 1/(3\sqrt{n}). [BK1997-lb-1-3sqrt]
0.30060.3006 [Kne2025] Lower bound for the bidisc: K2≥0.3006K_2\ge 0.3006. [Kne2025-lb-K2-0-3006]
References
  1. [1]Davis, Damek; Ivanisvili, Paata; Tao, Terence (maintainers). Optimization Constants in Mathematics, github.com/teorth/optimizationproblems (commit 2c1968c, 26 September 2026). — Bohr radius for the bidisc ($C_{59}$) linkApache-2.0 (repository license); constant pages reproduced with attribution; repository references retained.

Recorded progress

Add yours →

Nothing recorded yet. A pointer to a paper the entry is missing counts.

Machine attempts

0 recorded

Attempts run against the statement below and are recorded in full, including the failures. The point is to stop the next model — and the next person — from re-walking a road that is already known to end. Each attempt costs 25 credits.

No machine attempts recorded.

Sign in to run an attempt. New accounts start with enough credits for ten.

Discussion

0 notes

Notes are for clarifying the statement, flagging a missing reference, or recording why an approach fails. Claimed solutions belong in a submission, where they can be reviewed.

No notes yet.

Sign in to add a note.