betaFeedback

WorksMathematics

The inverse Littlewood problem for multiplicative functions

Abstract

We formulate and solve the asymptotic inverse Littlewood problem in the class of multiplicative functions. That is, let f:N→U∪{0}f:\mathbb{N}\to\mathbb{U}\cup\{0\} be a bounded completely multiplicative function satisfying ∑p:f(p)=01/p<∞\sum_{p:f(p)=0}1/p<\infty. We show that ∫01∣∑n≤xf(n)e(αn)∣ dα≪log⁡x \int_0^1 \left|\sum_{n\le x}f(n)e(αn)\right|\,dα\ll\log x for all x≥3x\ge3 if and only if f(n)=χ(n)nitf(n)=χ(n)n^{it} for some Dirichlet character χχ and some t∈Rt\in\mathbb{R}.\ More generally, we show that there is ε0>0\varepsilon_0>0, depending only on the ℓ2\ell^2-density of ff (that is, on ∏p:f(p)=0(1−1/p)\prod_{p:f(p)=0}(1-1/p)), such that under the weaker hypothesis ∫01∣∑n≤xf(n)e(αn)∣ dα≪xε0 \int_0^1 \left|\sum_{n\le x}f(n)e(αn)\right|\,dα \ll x^{\varepsilon_0} for all x≥3x\ge3, there are a Dirichlet character χχ and t∈Rt\in\mathbb{R} such that ∑p∣f(p)−χ(p)pit∣p<∞. \sum_p\frac{\left|f(p)-χ(p)p^{it}\right|}{p}<\infty. We give evidence for the sharpness of such classification, by constructing non-pretentious examples with L1L^1 norm o(x)o(\sqrt{x}) at every sufficiently large scale, but with ``wild" behavior at the intermediate scales.

Introduction

Throughout the paper, let U\mathbb{U} denotes the unit circle, and we write

e(u):=e2πiu.e(u):=e^{2\pi i u}.

Littlewood conjectured that there is an absolute constant c>0c>0 such that every finite set S⊂NS\subset\mathbb{N} satisfies

∫01∣∑n∈Se2πinα∣ dα>clog⁡∣S∣.(1)\int_0^1 \left|\sum_{n\in S}e^{2\pi i n\alpha}\right|\,\mathrm{d}\alpha>c\log|S|. \tag*{(1)}

He further conjectured that, among all S⊂NS\subset\mathbb{N} with ∣S∣=N|S|=N, the integral in (1) is smallest for S=[1,N]S=[1,N]. The first conjecture was proved independently by Konyagin [19] and by McGehee, Pigno and Smith [23], while the second one remains wide open. More generally, the results of [23] give the same lower bound for arbitrary coefficients a:N→Ca:\mathbb{N}\to\mathbb{C} with ∣a(n)∣≥1|a(n)|\geq1 for all n≥1n\geq1, that is

∫01∣∑n≤Na(n)e(nα)∣ dα≥clog⁡N\int_0^1 \left|\sum_{n\leq N}a(n)e(n\alpha)\right|\,\mathrm{d}\alpha\geq c\log N

for an absolute constant c>0c > 0. This lower bound is remarkably general, however not much is known about the extremizers. For coefficients of modulus one, generically we expect the left-hand side to be closer to the trivial upper bound N1/2N^{1/2}, which follows from the Cauchy–Schwarz inequality and observation that

∫01∣∑n≤Na(n)e(nα)∣2dα=∑n≤N∣a(n)∣2=N.\int_0^1 \left|\sum_{n\le N} a(n)e(n\alpha)\right|^2 d\alpha= \sum_{n\le N} |a(n)|^2 = N.

In general, the upper bound is achieved in many cases, in particular when SS is a lacunary set, or a random positive density subset of an interval. Interestingly, the only examples we know that achieve the lower bound O(log⁡N)O(\log N) are the unions of a bounded number of arithmetic progressions and a small unstructured set.

The inverse Littlewood problem was posed formally by Green in [12] (see also the paper of Petridis [28 Question 5.1] for a more precise formulation of the conjecture) and asks for a characterization of extremal sets SS for which this norm is O(log⁡∣S∣)O(\log|S|). Roughly speaking, we expect that

∫01∣∑n≤Na(n)e(nα)∣dα(2)\int_0^1 \left|\sum_{n\le N} a(n)e(n\alpha)\right|d\alpha \tag*{(2)}

is small when a(n)a(n) possesses some strong “additive” structure. See [9], [7], [28], [14], [13], [29] for some of the results in this direction.

In this paper, we confine ourselves to the classical framework in analytic number theory, requiring the sequence a(n)a(n) to be multiplicative. It is generally believed that multiplicative functions such as the Möbius function or the Liouville function possess no strong additive structure, and consequently, the L1L^1 norm (1.2) should be of the order ≫N1/2−o(1)\gg N^{1/2-o(1)}. For a≡μa\equiv\mu, lower bounds for (1.2) were obtained by Balog and Perelli [2] and by Balog and Ruzsa [3]; see also [4] for the closely related case of rr-free integers. More recently, Pandey and Radziwiłł [26] showed that for both a≡μa\equiv\mu and a≡λa\equiv\lambda the quantity (1.2) is ≫N1/4−o(1)\gg N^{1/4-o(1)}, making connection to the low lying zeros of Riemann ζ\zeta function. For a≡λa\equiv\lambda, this was improved to ≫N1/4\gg N^{1/4} by Li, Sun and Wang [22] and to Ω(N1/3−o(1))\Omega(N^{1/3-o(1)}) by Krause, Li, Sun and Wang [20]. Sharper quantitative bounds are known for other arithmetic weights, such as the indicator function of square-free numbers [10] and the Fourier coefficients of modular forms [25].

Our point of departure is a recent breakthrough result by Pandey and Radziwiłł [27] (see Theorem 1.5 below) showing that a multiplicative function whose exponential sum has a small L1L^1 norm must “pretend” to be a Dirichlet character at large primes. In particular, they established the following result, which concerns real valued functions ff.

Theorem 1.1 (Pandey–Radziwiłł [27]). Let f:N→[−1,1]f:\mathbb{N}\to[-1,1] be multiplicative and suppose that

lim inf⁡x→∞1x∑n≤xf(n)2>0.\liminf_{x\to\infty}\frac{1}{x}\sum_{n\le x} f(n)^2>0.

If

Nf(x):=∫01∣∑n≤xf(n)e(nα)∣dα=xox→∞(1)(x→∞),\mathcal{N}_f(x):=\int_0^1\left|\sum_{n\le x}f(n)e(n\alpha)\right|d\alpha=x^{o_{x\to\infty}(1)}\qquad(x\to\infty),

then there exists a real primitive Dirichlet character χ\chi such that

∑p≤x1−f(p)χ(p)p=o(∑p≤x1p)=o(log⁡log⁡x).\sum_{p\le x}\frac{1-f(p)\chi(p)}{p}=o\left(\sum_{p\le x}\frac{1}{p}\right)=o(\log\log x).

Clearly, for f:N→{−1,0,1}f:\mathbb{N}\to\{-1,0,1\} such a mean-square hypothesis is equivalent to ∑p:f(p)=01p<∞\sum_{p:f(p)=0}\frac{1}{p}<\infty.

Main results

Our aim in this paper is to obtain precise structural conclusions for a fixed completely multiplicative function from an L1L^1 bound that holds at every sufficiently large scale. The first result in this direction characterizes the functions for which the norm has the smallest possible logarithmic growth.

Theorem 1.2. Let f:N→U∪{0}f:\mathbb{N}\to\mathbb{U}\cup\{0\} be a completely multiplicative function satisfying ∑p:f(p)=01p<∞\sum_{p:f(p)=0}\frac{1}{p}<\infty (so that ∑n≤x∣f(n)∣2≫x\sum_{n\le x}|f(n)|^2\gg x) and

Nf(x)≪log⁡x(3)N_f(x)\ll\log x \tag*{(3)}

for all x≥3x\ge3. Then ff is of the form f(n)=χ(n)nitf(n)=\chi(n)n^{it}, where χ\chi is a Dirichlet character and t∈Rt\in\mathbb{R}.

One may view the conclusion of Theorem 1.2 as a strong manifestation of the asymptotic inverse Littlewood conjecture for the class of multiplicative functions. Interestingly, the converse also holds, in fact, it is easy to show that a function of the form f(n)=χ(n)nitf(n)=\chi(n)n^{it} satisfies (1.3). Our second main result provides an optimal improvement of Theorem 1.1, which also works for complex multiplicative functions.

Theorem 1.3. For every δ∈(0,1]\delta\in(0,1] there exists ε0=ε0(δ)>0\varepsilon_0=\varepsilon_0(\delta)>0, depending only on δ\delta, such that the following holds. Let f:N→Cf:\mathbb{N}\to\mathbb{C} be a completely multiplicative function with ∣f∣≤1|f|\le1, and suppose that

∑n≤x∣f(n)∣2≥δx(4)\sum_{n\le x}|f(n)|^2\ge\delta x \tag*{(4)}

for all sufficiently large xx. If

lim sup⁡x→∞log⁡Nf(x)log⁡x≤ε0(δ),(5)\limsup_{x\to\infty}\frac{\log N_f(x)}{\log x}\le\varepsilon_0(\delta), \tag*{(5)}

then there exist a primitive Dirichlet character χ\chi and t∈Rt\in\mathbb{R} such that

∑p∣f(p)−χ(p)pit∣p<∞.(6)\sum_p\frac{|f(p)-\chi(p)p^{it}|}{p}<\infty. \tag*{(6)}

The conclusion of Theorem 1.3 is significantly stronger than the statement that ff is merely “pretentious” (that is D(f,nitχ;∞)<∞\mathbb{D}(f,n^{it}\chi;\infty)<\infty). Moreover, the hypothesis (1.5) holds, in particular, if Nf(x)≪fxε0(δ)N_f(x)\ll_f x^{\varepsilon_0(\delta)} for all x≥3x\ge3 and crucially the implied constant may depend on ff in an arbitrary way since it only affects the size of the sum in (1.6) and not the constant ε0(δ)\varepsilon_0(\delta). Finally, if f:N→U∪{0}f:\mathbb{N}\to\mathbb{U}\cup\{0\} is completely multiplicative and ∑p:f(p)=01/p<∞\sum_{p:f(p)=0}1/p<\infty, then clearly

lim⁡x→∞1x∑n≤x∣f(n)∣2=∏p:f(p)=0(1−1p)>0,\lim_{x\to\infty}\frac{1}{x}\sum_{n\le x}|f(n)|^2=\prod_{p:f(p)=0}\left(1-\frac{1}{p}\right)>0,

so Theorem 1.3 applies with δ=12∏p:f(p)=0(1−1/p)\delta=\frac{1}{2}\prod_{p:f(p)=0}(1-1/p), say.

It is natural to ask if the hypothesis of Theorem 1.3 are necessary, in particular, whether one can seek similar characterization if lim sup⁡\limsup in (1.5) is replaced by a weaker information on various scales. In this direction, Theorem 1.4 below shows that lim sup⁡\limsup in (1.5) cannot be replaced by lim inf⁡\liminf, and so information about the behavior of Nf(x)N_f(x) on all large scales x≥3x\ge3 is required to obtain meaningful rigidity results.

References

  1. [1]A. Balog, A. Granville and K. Soundararajan, Multiplicative functions in arithmetic progressions, Ann. Math. Québec 37 (2013), 3–30; arXiv:math/0702389 (Theorem 2.1).
  2. [2]A. Balog and A. Perelli, On the L¹ mean of the exponential sum formed with the Möbius function, J. London Math. Soc. 57 (1998), 275–288.DOI
  3. [3]A. Balog and I. Z. Ruzsa, A new lower bound for the L¹ mean of the exponential sum with the Möbius function, Bull. London Math. Soc. 31 (1999), 415–418.DOI
  4. [4]A. Balog and I. Z. Ruzsa, On the exponential sum over r-free integers, Acta Math. Hungar. 90 (2001), 219–230.DOI
  5. [5]P. Bazin, Exponential sums over primes are unbounded, preprint (2025), arXiv:2508.18394.arxiv.org/abs/2508.18394
  6. [6]P. Bazin, I. Pylaev and F. Tyrrell, Bounded exponential sums with multiplicative coefficients, Mathematika 72 (2026), e70117.DOI
  7. [7]B. Bedert, Large sum-free subsets of sets of integers via L¹ estimates for trigonometric series, preprint (2025), arXiv:2502.08624.arxiv.org/abs/2502.08624
  8. [8]J. Benatar, A. Nishry and B. Rodgers, Moments of polynomials with random multiplicative coefficients, Mathematika 68 (2022), 191–216.DOI
  9. [9]T. Bloom and B. Green, Remarks on the Inverse Littlewood Conjecture, preprint (2026), arXiv:2602.16482.arxiv.org/abs/2602.16482
  10. [10]J. Brüdern, A. Granville, A. Perelli, R. C. Vaughan and T. D. Wooley, On the exponential sum over k-free numbers, Philos. Trans. Roy. Soc. London Ser. A 356 (1998), 739–761.DOI
  11. [12]B. Green, Approximate algebraic structure, Proceedings of the ICM, Seoul 2014, Vol. 1, 341–367.arxiv.org/abs/1404.0093
  12. [13]B. Green and T. Sanders, A quantitative version of the idempotent theorem in harmonic analysis, Ann. of Math. 168 (2008), 1025–1054.DOI
  13. [14]B. Hanson, Littlewood’s problem for sets with multidimensional structure, Int. Math. Res. Not. 2021, no. 21, 16736–16750.arxiv.org/abs/2003.01561
  14. [15]S. Hardy, Bounds for Exponential Sums With Random Multiplicative Coefficients, Int. Math. Res. Not. 2024, no. 22, 14138–14156.DOI
  15. [16]O. Klurman, Correlations of multiplicative functions and applications, Compositio Math. 153 (2017), no. 8, 1622–1657DOI
  16. [17]O. Klurman and A. P. Mangerel, Rigidity theorems for multiplicative functions, Math. Annalen 372 (2018), no. 1, 651–697arxiv.org/abs/1707.07817
  17. [18]O. Klurman, A. P. Mangerel and J. Teräväinen, On Elliott’s conjecture and applications, Proc. London Math. Soc. 133 (2026), no. 3, e70205.
  18. [19]S. V. Konyagin, On a problem of Littlewood, Math. USSR-Izv. 18 (1982), 205–225.
  19. [20]B. Krause, J. Li, Y. Sun and H. Wang, On the lower bound for the L¹ norm of Liouville exponential sums. II, preprint (2026), arXiv:2611.10790arxiv.org/abs/2611.10790
  20. [21]Y. Lamzouri, On the average of the number of imaginary quadratic fields with a given class number, Ramanujan J. 44 (2017), no. 2, 411–416.arxiv.org/abs/1512.07134
  21. [22]J. Li, Y. Sun and H. Wang, On the lower bound for the L¹ norm of Liouville exponential sums. I, preprint (2026), arXiv:2611.03589arxiv.org/abs/2611.03589
  22. [23]O. C. McGehee, L. Pigno and B. Smith, Hardy’s inequality and the L¹ norm of exponential sums, Ann. of Math. 113 (1981), 613–618.DOI
  23. [24]H. L. Montgomery and R. C. Vaughan, Mean values of multiplicative functions, Period. Math. Hungar. 43 (2001), no. 1–2, 199–214.DOI
  24. [25]M. Pandey, On the distribution of additive twists of the divisor function and Hecke eigenvalues, preprint (2021), arXiv:2110.03202.arxiv.org/abs/2110.03202
  25. [26]M. Pandey and M. Radziwiłł, L¹ means of exponential sums with multiplicative coefficients. I, preprint (2023), arXiv:2307.10329.arxiv.org/abs/2307.10329
  26. [27]M. Pandey and M. Radziwiłł, L¹ means of exponential sums with multiplicative coefficients. II, preprint (2025), arXiv:2510.20194.arxiv.org/abs/2510.20194
  27. [28]G. Petridis, The L¹-norm of exponential sums in ℤᵈ, Math. Proc. Cambridge Philos. Soc. 154 (2013), 381–392.DOI
  28. [29]S. Pichorides, On the L¹ norm of exponential sums, Ann. Inst. Fourier 30 (1980), 79–89.DOI
  29. [30]K. Soundararajan, Weak subconvexity for central values of L-functions, Ann. of Math. (2) 172 (2010), no. 2, 1469–1498.arxiv.org/abs/0809.1635
  30. [31]G. Tenenbaum, Crible d’Ératosthène et modèle de Kubilius, in: K. Győry, H. Iwaniec and J. Urbanowicz (eds.), Number Theory in Progress, Vol. 2 (Zakopane–Kościelisko, 1997), Walter de Gruyter, Berlin, 1999, 1099–1129.
  31. [33]R. C. Vaughan, The L¹ mean of exponential sums over primes, Bull. London Math. Soc. 20 (1988), 121–123.DOI

Paper details

vlnc:W_a2zecxnn
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

The AI tools

View statement
Looking for related papers…
Contents