Pith. sign in

REVIEW 5 major objections 4 minor 38 references

A generalization of Littlewood's $L^\alpha$ flat theorem, $\alpha>0$

T0 review · 5 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

Pith's one-line read The paper proves that no normalized analytic polynomial sequence in Littlewood's coefficient class is L^p-flat for any p>0, settling the L^α-Littlewood, Newman, and Erdős conjectures.

desk verdict A serious attempt at a major result, but the main proof has a fatal gap: Lemma 3 is false and the L^alpha to L^r step is unjustified. read the letter →

arxiv 2509.04212 v1 pith:RVWB6G26 submitted 2025-09-04 math.NT math.DS

classification math.NTmath.DS MSC 11C0842A0542A5537A0537A3042A61
keywords flatpolynomialsultraflatErdős–LittlewoodproblemunimodularClarksoninequalitiesBarkersequencesgeneralizedRieszproductssingularspectrum
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proves that within a broad class of analytic polynomials on the unit circle—those whose coefficients satisfy Littlewood's growth condition, where the unweighted sum of squared coefficients is bounded by K/n² times the frequency-weighted sum—no sequence of normalized polynomials can be flat in any L^p sense for p>0. The proof takes Littlewood's classical norm-gap criterion for real trigonometric polynomials and transfers it to complex polynomials by splitting each polynomial into real and imaginary parts and applying a generalized Clarkson inequality for Banach-space-valued L^p functions. If correct, this settles the L^α-Littlewood conjecture, and with it Newman's L¹ conjecture and Erdős's L^∞ conjecture. It also implies that only finitely many Barker sequences exist, that certain Morse cocycles have singular spectra, and that Gauss–Fresnel polynomials contain Mahler-flat subsequences, giving a new proof of the Beller–Newman theorem.

What carries the argument

The key mechanism is the Littlewood coefficient condition Σ a_m² ≤ K/n² Σ m² a_m², which controls the derivative of the trigonometric polynomial and yields a uniform gap between its L^α and L² norms. The proof represents the normalized analytic polynomial as f̃_n = (g̃_n + i h̃_n)/√2, where g̃_n and h̃_n are its real and imaginary parts, each of L² norm 1, and applies the generalized Clarkson second inequality for Banach-valued L^α spaces to the pair F = g̃_n/√2 and G = i h̃_n/√2. This convexity inequality controls the L^r norm of f̃_n in terms of the L^s norms of its parts; feeding in Littlewood's norm gap and letting the exponent r tend to 2 from above yields the contradiction that establi

What would settle it

Take any fixed unimodular sequence c_j and any fixed real sequence a_j satisfying Σ a_m² ≤ K/n² Σ m² a_m², and compute the normalized L^p deviation (∫_{S¹} ||P_n(z)|-1|^p dz)^{1/p} for increasing n. The theorem predicts the limsup of this quantity is positive for every p>0; finding one admissible sequence with this deviation tending to zero for any p would falsify the central claim.

Watch

Extended reading notes

Core claim

The paper's central claim is that flatness is impossible, at every exponent p>0, for normalized analytic polynomials P_n(z) = (Σ a_j c_j z^j)/(Σ a_j²)^{1/2} whenever the real coefficients a_j obey Σ a_m² ≤ K/n² Σ m² a_m² for an absolute constant K and the c_j are unimodular. Littlewood had established norm gaps for real cosine and sine polynomials under this condition; the new step is to transfer that gap to the modulus of the complex polynomial by writing its real and imaginary parts as two real polynomials of equal norm and applying a Clarkson-type inequality to the pair. The argument derives a contradiction: if the sequence were L^α-flat, the Clarkson inequality together with Littlewood's

Load-bearing premise

The proof depends on the assumption that closeness to 1 in one average sense implies closeness to 1 in a slightly stronger average sense; this is false in general because large values on very small sets can inflate the stronger average without affecting the weaker one.

Editorial extensions

If this is right

  • Littlewood polynomials with coefficients ±1 are not L^α-flat for any α>0, confirming the L^α-Littlewood conjecture.
  • Newman's L¹ conjecture and Erdős's L^∞ conjecture follow, since a flat sequence at either exponent would fall inside the forbidden class.
  • Uniform unimodular polynomials—fixed coefficient sequences with |c_j|=1—are never ultraflat.
  • There are only finitely many Barker sequences, because an infinite family would produce L²-flat Littlewood polynomials.
  • Morse cocycles with ±1 values over an odometer have singular maximal spectral type, resolving an old ergodic-theory question through Guenais's criterion.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The theorem's scope ends where coefficients depend on the degree: Newman's Gauss-sum polynomials and Kahane-type ultraflat constructions escape the coefficient condition precisely because their coefficients are not fixed in advance, so their flatness is not contradicted by this result.
  • Because Littlewood's criterion is a limsup statement, the proof should also rule out L^p-flatness for coefficient sequences satisfying the condition along sparse subsequences of degrees, not only for every n.
  • A quantitative refinement that makes the constant A(K,α) explicit would turn the theorem into effective lower bounds on the L^α deviation of finite Littlewood polynomials; the paper notes the constant is not yet computed.
  • The proof template—splitting a complex function into equal-norm real and imaginary parts and applying a uniform-convexity inequality—could be tested on random coefficient models satisfying the same weighted-square condition.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. The paper claims a generalization of Littlewood's L^α-flatness criterion: no sequence of normalized analytic polynomials P_n(z) = (∑_{j=1}^n a_j c_j z^j)/(∑ a_j^2)^{1/2} with ∑ a_m^2 ≤ (K/n^2) ∑ m^2 a_m^2 can be L^p-flat for any p>0. From this it derives several corollaries, including the L^α-Littlewood conjecture, the L^1-Newman and L^∞-Erdős conjectures, finiteness of Barker sequences, singularity of certain Morse cocycle spectra, and a new proof of the Beller–Newman theorem via Gauss–Fresnel polynomials. The proof of the main theorem combines Littlewood's L^α criterion with a generalized Clarkson inequality (Lemma 3) and a contradiction argument letting a parameter r approach 2.

Significance. If the main theorem were correct, it would indeed resolve a family of long-standing conjectures in one stroke, and the applications to ergodic theory and number theory would be notable. The paper also includes some original ideas: the reduction of the problem to real and imaginary parts and the use of generalized Clarkson inequalities is a plausible route. However, the central proof contains a serious, load-bearing gap: the deduction from L^α-flatness to convergence of higher L^r norms is not justified and is false in general. Moreover, the stated Lemma 3 is itself invalid, and the passage from the fixed-sequence Theorem 1 to n-dependent Littlewood coefficients is not covered. These issues are not local; they affect the main result and its advertised consequences.

major comments (5)
  1. [§1, proof of Theorem 1, eqs. (11)–(14)] The step from (13) to (14) assumes that ‖f̃_n‖_r → 1 for r = 2+δ > α under L^α-flatness. The hypothesis only gives ‖f̃_n‖_α → 1. L^α convergence does not control higher L^r moments; a sequence with spikes on sets of small measure illustrates the failure. Without an additional uniform bound on ‖f̃_n‖_r, the left-hand side of (11) need not converge to 2^{1/r}, and the contradiction does not follow.
  2. [§1, Lemma 3] Lemma 3 as stated is false. Taking F≡1, G≡0 on a probability space with s=p=α and r=2 gives LHS=2^{1/2} and RHS=2^{1-1/α}; for α<2 the inequality fails. The inequality is also non-homogeneous: the LHS scales as λ^{p/r} while the RHS scales as λ^{p/s}. Moreover, the parameter condition r′≤s≤p forces r≥α′=α/(α−1)>2 for α∈(1,2), so the final 'letting δ→0' in (14)–(16) is unavailable for fixed α. Thus the chain (11)–(16) cannot produce the claimed contradiction.
  3. [§1, Corollary 3] Theorem 1 is stated for a fixed infinite coefficient sequence (a_j), (c_j). Corollary 3 applies it to polynomials whose coefficients ε_n(k) depend on n. This is not covered by the theorem as stated; a diagonal argument or a uniform version for n-dependent coefficients is missing. The advertised consequences (L^α-Littlewood, Newman, Erdős) concern sequences of Littlewood polynomials with signs varying with n, so this gap is load-bearing for the abstract's main claim.
  4. [§5, proof of Theorem 3] The displayed identity ‖ |P_m|−1 ‖_2^2 = 2 − ∫|P_m| dz is incorrect; the right-hand side should be 2 − 2∫|P_m| dz. More substantively, Lemma 9 produces a subsequence of stretched polynomials P_{j_k}(z^{l_k}) with a product finite and nonzero a.e., and Lemma 10 gives M(µ)=∏ M(P_{j_k})^2. Neither implies that M(P_m) → 1 for the original Gauss–Fresnel polynomials. The alternative proof of the Beller–Newman theorem is therefore incomplete.
  5. [§3, Corollary 5] The bounds involving √N do not follow from Theorem 1 or Lemma 1 for the analytic polynomial norm. Littlewood's criterion gives information about the real cosine polynomial g_N, not |P_N|. For α<2, the inequality ‖g_N‖_α ≤ (1−A)‖g_N‖_2 does not imply ‖P_N‖_α ≤ (1−C)√N, since |P_N| ≥ |g_N|. The α>2 case similarly compares to √(N/2), not √N. The statement needs a justification or reformulation.
minor comments (4)
  1. [Abstract/§5] Numerous typos and misspellings: 'Furstermore', 'generalizatin', 'anwser', 'unimodulair', 'polynomails', 'Frenesl' for 'Fresnel', 'wearker', 'Baker sequences' for 'Barker sequences', and 'conclude the proof' repeated. The paper would benefit from a careful proofreading.
  2. [§1, proof of Corollary 3] The line “Littlewood's argument |P_n(z) − P_n(z′)| = ±{|P_n(z)|} ± {|P_n(z′)|}” is not meaningful as written and does not explain how Littlewood's method applies.
  3. [§1, Lemma 4] The notation ζ2 is confusing and seems to denote a constant rather than a function of ζ; the bound “|ζ|,|ζ2| < BεA” is ambiguous. Consider restating with clearer constants.
  4. [§2] “discret” should be “discrete”, and the statement that L^α-flatness implies ultraflatness (given in the introduction) is incorrect as written: L^α-flatness only gives convergence in L^α of |Q_n|−1, not uniform convergence.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central theorem is derived from external Littlewood and Boas/Clarkson inequalities; self-citations appear mainly in applications and are not load-bearing.

full rationale

The central derivation is not circular. Theorem 1 is proved by contradiction from external classical results—Littlewood's L^α criterion (Lemma 1, ref [28]), its sine-polynomial variant (Lemma 2), and the generalized Clarkson inequality (Lemma 3, refs [12],[26])—none of which is defined in terms of the theorem's conclusion. The coefficient condition in Theorem 1 is exactly the hypothesis of Littlewood's criterion, so the proof transfers that external criterion from real to complex polynomials; it does not assume what it proves. Self-citations ([1],[2],[4],[6]) occur in applications and background: the prior α≥4 result, generalized Riesz-product calculus, and palindromic Littlewood polynomials. Lemma 10 of [4] is a general parameter-free fact about Mahler measures of Riesz products and does not presuppose the Beller–Newman theorem it helps reprove. Hence no fitted input, self-citation load-bearing step, uniqueness import, or renaming is present in the main theorem. I do flag, as non-circular correctness issues (per the review rule), that Lemma 3 as printed appears non-homogeneous and misapplied with r=2+δ→2 at equations (14)–(16), and that Corollary 3 uses n-dependent coefficient sequences not covered by Theorem 1's fixed-sequence hypothesis; these flaws undermine the proof but are not circular reductions.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The central claim rests on classical external theorems (Littlewood's L^α criterion, Boas's generalized Clarkson inequality, Turyn-Storer, and Guenais) and on two lemmas from the author's earlier work with Nadkarni ([4]) used in the Mahler-flatness section. No new entities or fitted parameters are introduced. The proof does not appear circular in the sense of assuming the target result; the flaw is an unjustified norm-convergence step, not circularity.

assumptions (7)
  • standard math Littlewood's L^α-flatness criterion (Lemma 1, Section 1): for real trigonometric polynomials g_n with coefficient condition, limsup ||g_n||_α/||g_n||_2 ≤ 1−A(K,α) for α<2 and ≥ 1+A(K,α) for α>2.
    Core estimate used in the proof of Theorem 1 to bound the L^α norms of the real and imaginary parts of the polynomial.
  • standard math Boas's generalized second Clarkson inequality for L^p(X; B) (Lemma 3, equation (2)). The printed statement is not homogeneous and as written cannot be true for r ≠ s.
    Invoked in Theorem 1 to compare the L^r norm of the sum F+G with L^s norms of F and G; the misstatement is a red flag for the validity of the proof.
  • standard math WLOG reduction to 0 < α < 2: L^α-flatness for α>2 implies L^β-flatness for β<α by Hölder.
    Used at the start of the proof of Theorem 1 to assume 1<α<2 so that Boas's inequality applies.
  • domain assumption Turyn-Storer theorem (Theorem 2(1), Section 4): odd-length Barker sequences have length ≤ 13, even length > 2 must equal 4m^2.
    Used with the main theorem to derive Corollary 6, the claim that there are only finitely many Barker sequences.
  • domain assumption Guenais's theorem (Section 2, [21]): for the Morse cocycle, the spectral measure σ_φ is singular if the associated Littlewood polynomials are not L^1-flat.
    Used to convert the main theorem's non-L^1-flatness conclusion into singularity of the spectral type in Corollary 4.
  • domain assumption Lemmas 9 and 10 from the author's earlier work with Nadkarni [4] on generalized Riesz products and their Mahler measures.
    Used in the proof of Theorem 3 to conclude Mahler-flatness of a subsequence of Gauss-Fresnel polynomials; these are self-cited results.
  • ad hoc to paper L^α-flatness implies ||f̃_n||_α → 1 and also, implicitly in the proof, ||f̃_n||_r → 1 for r > α; the latter is false without further assumptions.
    The paper's proof silently assumes the L^r norm convergence, which is the central unjustified step.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A generalization of Littlewood's $L^\alpha$ flat theorem, $\alpha>0$." pith.science (2026). https://pith.science/paper/RVWB6G26

@misc{pith2026250904212,
  author       = {Pith},
  title        = {Pith review of: A generalization of Littlewood's $L^\alpha$ flat theorem, $\alpha>0$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RVWB6G26}},
  note         = {Machine review of arXiv:2509.04212}
}
abstract

We establish a generalization of Littlewood's criterion on $L^\alpha$-flatness by proving that there is no $L^\alpha$-flat polynomials, $\alpha>0$, within the class of analytic polynomials on the unit circle of the form $ P_n(z)=\sum_{m=1}^{n}c_m z^m, n \in {\mathbb{N}}^*,$ satisfying $$ \sum_{m=1}^{n}|c_m|^2 \leq \frac{K}{n^2} \sum_{m=1}^{n}m^2 |c_m|^2, $$ where $K$ is an absolutely constant. As a consequence, we confirm the $L^\alpha$-Littlewood conjecture, and thereby the $L^1$-Newman and $L^\infty$-Erd\"os conjectures. Our approach combines the $L^\alpha$ Littlewood theorem with the generalized Clarkson's second inequality for $L^\alpha(X,\mathcal{A},m;B)$, with $B$ a Banach spaces and $1 < \alpha \leq 2.$ It follows that there are only finitely many Barker sequences, and we further present several applications in number theory and the spectral theory of dynamical systems. Finally, we construct Gauss-Fresnel polynomials that are Mahler-flat, providing a new proof of the Beller-Newman theorem.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages

  1. [1]

    e. H. el Abdalaoui, On Lp-flatness of Erd˝ os-Littlewood’s polynomials, arXiv:2504.21499 [math.CA], https://arxiv.org/abs/2504.21499

  2. [2]

    e. H. el Abdalaoui & M.G. Nadkarni, Some notes on flat polynomi- als,arXiv:1402.5457 [math.CV],http://arxiv.org/abs/1402.5457

  3. [3]

    e. H. el Abdalaoui & M. G. Nadkarni, On flat polynomials with non-negative coefficients, arXiv:1508.00417v1 [math.DS], https://arxiv.org/abs/1508. 00417

  4. [4]

    e. H. el Abdalaoui & M. G. Nadkarni, Calculus of generalized Riesz products, Contemp. Math., 631, 145-180

  5. [5]

    e. H. el Abdalaoui, Exponential sums of the M¨ obius function and flat polyno- mials, arXiv:1910.10569 [math.NT]

  6. [6]

    H., A Class of Littlewood Polynomials that are Not Lα-Flat ( with an appendix with M

    el Abdalaoui, e. H., A Class of Littlewood Polynomials that are Not Lα-Flat ( with an appendix with M. G. Nadkarni), Unif. Distrib. Theory 15 (2020), no.1, 51-74

  7. [7]

    e. H. el Abdalaoui & M. G. Nadkarni, A non-singular transformation whose spectrum has Lebesgue component of multiplicity one, Erg. Theo. Dyn. Sys., 36 (2016), pp 671-681

  8. [8]

    Alexanderson, G. L. & P. Erd˝ os. An Interview with Paul Erd˝ os. The Two- Year College Mathematics Journal 12, no. 4 (1981): 249–59. https://doi. org/10.2307/3027072

Show all 38 references
  1. [9]

    R. H. Barker, Group synchronizing of binary digital systems, Communication Theory (London, 1952) (W. Jackson, ed.), Academic Press, New York, 1953, pp. 273–287

  2. [10]

    Baker and G

    R.C. Baker and G. Harman, Exponential sums formed with the M¨ obius func- tion. J. London Math. Soc. 43 (1991) 193-198

  3. [11]

    Beller and D

    E. Beller and D. J. Newman, An Extremal problem for the geometric mean of polynomials, Proc. Amer. Math. Soc., Vol. 39, No. 2 (Jul., 1973), pp. 313-317

  4. [12]

    R. P. Jr. Boas, Some uniformly convex spaces. Bull. Amer. Math. Soc. 46, (1940), 304- 311. 26 el Houcein el Abdalaoui ⋆

  5. [13]

    Borwein and M

    P. Borwein and M. J. Mossinghoff, Barker sequences and flat polynomials, Number Theory and Polynomials (Bristol, U.K., 2006) (J. McKee and C. Smyth, eds.), London Math. Soc. Lecture Note Ser., vol. 352, Cambridge Univ. Press, 2008, pp. 71–88

  6. [14]

    Borwein & M.J

    P. Borwein & M.J. Mossinghoff, Wieferich pairs and Barker sequences, II. LMS Journal of Computation and Mathematics. 2014;17(1):24-32. doi:10.1112/S1461157013000223

  7. [15]

    M. M. Day, Some more uniformlyconvex spaces. Bull. Amer. Math. Soc. 47 (1941), 504- 507

  8. [16]

    Erd´ elyi, T.,The sequence of partial sums of a unimodular power series is not ultraflat, arXiv:2504.19336v1 [math.NT], https://arxiv.org/abs/2504.19336

  9. [17]

    Erd˝ os, An inequality for the maximum of trigonometric, Ann

    P. Erd˝ os, An inequality for the maximum of trigonometric, Ann. Polon. Math., (12) 1962, 151–154. polynomials

  10. [18]

    Erd˝ os, Some unsolved problems, Michigan Math

    P. Erd˝ os, Some unsolved problems, Michigan Math. J., (4) 1957, 291-300

  11. [19]

    Erd˝ os, Problems and results on polynomials and interpolation

    P. Erd˝ os, Problems and results on polynomials and interpolation. Aspects of contemporary complex analysis (Proc. NATO Adv. Study Inst., Univ. Durham, Durham, 1979), pp. 383–391, Academic Press, London-New York, 1980

  12. [20]

    A. Y. Fawaz, On an unsolved problem in the analytic theory of numbers, Quart. J. Math., Oxford Ser. 3 (1952), 282–295

  13. [21]

    Guenais, Morse cocycles and simple Lebesgue spectrum, Ergodic Theory Dynam

    M. Guenais, Morse cocycles and simple Lebesgue spectrum, Ergodic Theory Dynam. Systems, 19 (1999), no. 2, 437-446

  14. [22]

    Hajela and B

    D. Hajela and B. Smith, On the maximum of an exponential sum of the M¨ obius function. Number theory (New York, 1984-1985), 145-164, Lecture Notes in Math., 1240, Springer, Berlin, 1987

  15. [23]

    Helson, Cocycles on the circle , J

    H. Helson, Cocycles on the circle , J. Operator Theory 16 (1986), no. 1, 189–199

  16. [24]

    Jedwab, A survey of the merit factor problem for binary sequences, In sequences and their applications-SETA 2004 (pp

    J. Jedwab, A survey of the merit factor problem for binary sequences, In sequences and their applications-SETA 2004 (pp. 30-55), Springer Berlin Hei- delberg

  17. [25]

    Kahane, Sur les polynˆ omes ` a coefficients unimodulaires,(French) Bull

    J-P. Kahane, Sur les polynˆ omes ` a coefficients unimodulaires,(French) Bull. London Math. Soc., 12 (1980), no. 5, 321-342

  18. [26]

    Koskela, Some generalizations of Clarkson’s inequalities,Publikacije Elek- trotehnickog Fakulteta

    M. Koskela, Some generalizations of Clarkson’s inequalities,Publikacije Elek- trotehnickog Fakulteta. Univerzitat u Beogradu. Serija Matematika i Fizika, Publ. Elektroteh. Fak., Univ. Beogr., Ser. Mat. Fiz., Vol (634-677),89-93,1979

  19. [27]

    The London meetings 2012–2014

    Lawrence, S., What are we like · · ·, Larvor, Brendan (ed.), Mathematical cul- tures. The London meetings 2012–2014. Basel: Birkh¨ auser/Springer. Trends Hist. Sci., 329-346 (2016)

  20. [28]

    J. E. Littlewood, On the mean values of certain trigonometric polynomials. , J. London Math. Soc. 36, 1961, 307–334

  21. [29]

    M. G. Nadkarni, Spectral Theory of Dynamical Systems , Hindustan Book Agency, New Delhi, (1998); Birkh¨ auser Advanced Texts : Basler LehrbÆcher. A GENERALIZATION OF LITTLEWOOD’SLα-FLAT THEOREM 27 [Birkh¨ auser Advanced Texts: Basel Textbooks] Birkh¨ auser Verlag, Basel, 1998

  22. [30]

    Newman, Norms of polynomials, Amer

    D.J. Newman, Norms of polynomials, Amer. Math. Monthly, Vol. 67, No. 8 (Oct., 1960), pp. 778-779

  23. [31]

    D. J. Newman, An L1 extremal problem for polynomials, Proc. Amer. Math. Soc., Vol 16 (1965), 1287-1290

  24. [32]

    A. M. Odlyzko and H. J. J. te Riele, Disproof of the Mertens conjecture, J. Reine Angew. Math. 357 (1985), 138-160

  25. [33]

    In: Butler S, Cooper J, Hurlbert G, eds

    Odlyzko, A., Search for Ultraflat Polynomials with Plus and Minus One Co- efficients. In: Butler S, Cooper J, Hurlbert G, eds. Connections in Discrete Mathematics: A Celebration of the Work of Ron Graham. Cambridge Uni- versity Press; 2018:39-55

  26. [34]

    Saffari, Barker sequences and Littlewood two-sided conjectures on polyno- mials with ±1 coefficients, S´ eminaire d’Analyse Harmonique, Ann´ ee 1989/90, 139–151, Univ

    B. Saffari, Barker sequences and Littlewood two-sided conjectures on polyno- mials with ±1 coefficients, S´ eminaire d’Analyse Harmonique, Ann´ ee 1989/90, 139–151, Univ. Paris XI, Orsay, 1990

  27. [35]

    Turyn and J

    R. Turyn and J. Storer, On binary sequences, Proc. Amer. Math. Soc. 12 (1961), 394–399

  28. [36]

    E. C. Titchmarsh, The theory of the Riemann zeta-function, Second edition. Edited and with a preface by D. R. Heath-Brown. The Clarendon Press, Ox- ford University Press, New York, 1986

  29. [37]

    Yu, A Note on Barker Sequences and the L1-norm of Littlewood Poly- nomials, Comptes Rendus

    G. Yu, A Note on Barker Sequences and the L1-norm of Littlewood Poly- nomials, Comptes Rendus. Math´ ematique, Volume 361 (2023), pp. 609-616, url=https://doi.org/10.5802/crmath.428

  30. [38]

    Zygmund, Trigonometric Series vol

    A. Zygmund, Trigonometric Series vol. I & II, corrected second ed., Cam- bridge Univ. Press, Cambridge, 1968. ⋆ University of Rouen Normandy , Department of Mathematics, LMRS UMR 60 85 CNRS, A venue de l’Universit´e, BP.12 76801 Saint Etienne du Rouvray - France . Email addres...

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.