betaFeedback

WorksMathematics

Random Configurations and Fourier Decoupling

Abstract

We use tools from high dimensional probability to analyze random configurations of balls in the context of Fourier decoupling and we answer a question on the structure of superlevel sets in the setting of decoupling due to Guth.

Introduction

In this article we will study random configurations in the context of Fourier decoupling theory and will answer a question of Guth [13]. Heuristics coming from probability theory often play an important guiding role in Fourier analysis. The notions of square root cancellation, pseudo-randomness and probabilistic models for various Fourier-analytic and number theoretic objects can often tell us how these objects should behave. These heuristics can help us formulate conjectures and could also help us determine the types of inquiry that may yield progress on important problems. With this in mind, it can be fruitful to study random configurations in the context of Fourier analysis. For example, the work of Bourgain and Talagrand on Λ(p)\Lambda(p) sets [1] [19], and the work of Mockenhaupt and Schlag [17] on the Hardy-Littlewood majorant problem have helped govern heuristics for the study of exponential sums and have also found applications in compressed sensing and signal processing (see [16], [5], [21] for illustrative examples). We investigate ideas along these lines in the context of Fourier decoupling. A central question in this context is: what is the typical behavior of natural Fourier analytic objects? In [18] the author investigated random configurations in the context of the Mizohata-Takeuchi conjecture, and showed that the Mizohata-Takeuchi conjecture is true most1 of the time. We use some of the tools developed in [18] and refine them to answer questions about the typical behavior of configurations that naturally arise in the context of Fourier decoupling.

Setting of the Problem and Main Questions: We recall the setting of Fourier decoupling below. Let the set N1/N(S1)\mathcal{N}_{1/N}(S^1) be the 1/N1/N neighborhood of the circle S1⊂R2S^1 \subset\mathbb{R}^2. Divide N1/N(S1)\mathcal{N}_{1/N}(S^1) into NN approximate rectangles of dimensions 1/N×1/N21/N \times1/N^2 and call this collection {θ}\{\theta\}. Let f∈S(R2)f \in\mathcal{S}(\mathbb{R}^2) (the Schwartz class on R2\mathbb{R}^2) be such that

f=∑θfθwithsupp⁡(fθ^)⊆θ,(1)f = \sum_{\theta} f_\theta\quad\text{with} \quad\operatorname{supp}(\widehat{f_\theta}) \subseteq\theta, \tag*{(1)}

then a celebrated theorem of Bourgain and Demeter [2] states that

∥f∥L6(R2)≲ϵNϵ(∑θ∥fθ∥L6(R2)2)1/2.(2)\|f\|_{L^6(\mathbb{R}^2)} \lesssim_\epsilon N^\epsilon\left(\sum_{\theta}\|f_\theta\|_{L^6(\mathbb{R}^2)}^2\right)^{1/2}. \tag*{(2)}

Suppose in addition that for all θ\theta we also have that

∥fθ∥L∞(R2)≤1.\|f_\theta\|_{L^\infty(\mathbb{R}^2)} \leq1.

A consequence of more refined work by Guth, Maldague, and Wang on the decoupling problem yields a bound for the size of the super level sets UλU_\lambda. In this setting, define Uλ:={x∈R2:∣f(x)∣>λ}U_\lambda:= \{x \in\mathbb{R}^2 : |f(x)| > \lambda\}. Then if BN2:={x:∣x∣≤N2}B_{N^2} := \{x : |x| \le N^2\} (that is the ball of radius N2N^2 in R2\mathbb{R}^2), then in [15] it was shown that

∣Uλ∩BN2∣≲ϵNϵN7λ6.(3)|U_\lambda\cap B_{N^2}| \lesssim_\epsilon N^\epsilon\frac{N^7}{\lambda^6}. \tag*{(3)}

In light of this bound we are led to the following natural questions:

Questions:

  1. For different values of λ\lambda, what kinds of functions ff saturate the bound (3)?

  1. What can we say about the structure of UλU_\lambda if the bound (3) is saturated?

One can think of these questions as asking for an ‘inverse theorem’ in the context of Fourier decoupling. Inverse theorems have played a significant role in recent developments in additive combinatorics and number theory (see this excellent talk of Gowers [8] for more background), and a satisfactory answer to these questions could lead to important insights about the behavior of exponential sums and related questions (see this survey [11] of Guth for more discussion).

When λ=N1/2\lambda= N^{1/2} it is easy to find many examples where the bound (3) is saturated. On the other hand if N1/2<λ≤NN^{1/2} < \lambda\le N, there are very few examples of functions that saturate the bounds in (3), and it is not clear whether this bound is sharp for all values of λ\lambda. We remark that the question of the sharpness of (3) was investigated in [7], but in the case where S1S^1 was replaced by the parabola P1P^1, where the authors were able to leverage techniques coming from the study of some well known exponential sums. These techniques do not carry over to the case of the circle S1S^1. In [13] Guth asked whether the bound (3) can be saturated by a random set. In particular he asked the following:

  1. If VV is a union of N7λ−6N^7\lambda^{-6} random unit balls, and λ>N1/2\lambda> N^{1/2}, is it typically possible (in the setting above) that ∥f∥L1(V)∼λ∣V∣\|f\|_{L^1(V)} \sim\lambda|V|?

If the answer to this question was yes, then it would be that in the case when (3) was saturated, the set Uλ∩BN2U_\lambda\cap B_{N^2} could look like a random set. However, we answer this question in the negative. In fact we show that on the random set VV, ff will be strictly smaller than what the superlevel set bound (3) guarantees. To set up our main theorem we first define the class of functions F\mathcal{F} as follows:

F:={f∈S(R2):f=∑θfθ,supp⁡(fθ^)⊂θ,∥fθ∥L∞(R2)≤1}.(4)\mathcal{F} := \left\{ f \in\mathcal{S}(\mathbb{R}^2) : f = \sum_\theta f_\theta,\quad\operatorname{supp}(\widehat{f_\theta}) \subset\theta,\quad\|f_\theta\|_{L^\infty(\mathbb{R}^2)} \le1 \right\}. \tag*{(4)}

where the collection {θ}\{\theta\} is defined as in (1). Our main theorem gives a precise asymptotic for sup⁡f∈F∫V∣f(x)∣dx\sup_{f\in\mathcal{F}} \int_V |f(x)|dx.

Theorem 1 (Corollary of Main Theorem, Informal Statement). In the setting above, let VλV_\lambda be a random collection of N7λ−6N^7\lambda^{-6} unit balls contained in BN2B_{N^2}, then with high probability

sup⁡f∈F∥f∥L1(Vλ)∼ϵ{N1/2∣V∣when N1/2<λ≤N5/6,λ3N2∣V∣when N5/6≤λ≤N(5)\sup_{f\in\mathcal{F}} \|f\|_{L^1(V_\lambda)} \sim_\epsilon \begin{cases} N^{1/2}|V| & \text{when } N^{1/2}<\lambda\le N^{5/6},\\ \frac{\lambda^3}{N^2}|V| & \text{when } N^{5/6}\le\lambda\le N \end{cases} \tag*{(5)}

In any case one can deduce that whenever N1/2≤λ<NN^{1/2} \le\lambda< N, that

sup⁡f∈F∥f∥L1(Vλ)∼o(λ)∣Vλ∣.(6)\sup_{f\in\mathcal{F}}\|f\|_{L^1(V_\lambda)} \sim o(\lambda)|V_\lambda|. \tag*{(6)}

What this shows is that a set UλU_\lambda that saturates (3) cannot look random. From here, one can deduce that examples of sets Uλ∩BN2U_\lambda\cap B_{N^2} that saturate (3) must possess some kind of special structure. Finding examples of such sets and determining what kind of structure Uλ∩BN2U_\lambda\cap B_{N^2} must possess remains an important challenge. See [11] for related discussion.

The two different regimes for the estimate (5) may appear mysterious to the reader. We clarify this by proving a more general estimate for collections of random sets of arbitrary cardinality.

Theorem 2 (Main Theorem, Informal Statement). For each N≤q≤CN4N \le q \le CN^4, let VV be a random collection of qq unit balls contained inside BN2B_{N^2}, then with high probability

sup⁡f∈F∥f∥L1(V)∼εNε(N1/2q+N3/2q).(7)\sup_{f\in\mathcal{F}}\|f\|_{L^1(V)} \sim_\varepsilon N^\varepsilon\left(N^{1/2}q+N^{3/2}\sqrt{q}\right). \tag*{(7)}

when q=N7λ−6q=N^7\lambda^{-6}, one can check that the first term in (7) dominates whenever N1/2≤λ≤N5/6N^{1/2} \le\lambda\le N^{5/6} and then second term in (7) dominates when N5/6<λ≤NN^{5/6}<\lambda\le N. These observations yield the estimate (5).

Setup for Theorem 2

In this section we will convert the informal statement of Theorem 2 into a formal statement. Decompose BN2B_{N^2} into a finitely overlapping union of CN4CN^4 unit balls {Vk}k=1CN4\{V_k\}_{k=1}^{CN^4} for some absolute constant CC. For each 1≤k≤CN41 \le k \le CN^4, define the Bernoulli random variable δk\delta_k as follows:

P(δk=1)=O(qN4)=:δ,P(δk=0)=1−δ.(8)\mathbb{P}(\delta_k=1)=O\left(\frac{q}{N^4}\right)=:\delta,\qquad\mathbb{P}(\delta_k=0)=1-\delta. \tag*{(8)}

Then with high probability, the random set VV defined as

V:={⋃k:δk=1Vk}⊆BN2,(9)V:=\left\{\bigcup_{k:\delta_k=1}V_k\right\}\subseteq B_{N^2}, \tag*{(9)}

is such that ∣V∣∼q|V|\sim q.

Theorem 3. In the setting coming from the discussion above,

Esup⁡f∈F∥f∥L1(V)=Esup⁡f∈F∑k=1CN4δk∫Vk∣f(x)∣ dx∼εNε(N1/2q+N3/2q).(10)\mathbb{E}\sup_{f\in\mathcal{F}}\|f\|_{L^1(V)} =\mathbb{E}\sup_{f\in\mathcal{F}}\sum_{k=1}^{CN^4}\delta_k\int_{V_k}|f(x)|\,dx \sim_\varepsilon N^\varepsilon\left(N^{1/2}q+N^{3/2}\sqrt{q}\right). \tag*{(10)}

The first equality can easily be seen to be true by chasing definitions, the second inequality is the main technical result of this article.

The lower bound in (10) comes from some probabilistic constructions involving wave packets, which we show in § 7. The technical heart of this paper, however, is in proving the upper bound which we restate here.

Proposition 4. We have the following inequality:

Esup⁡f∈F∑k=1CN4δk∫Vk∣f(x)∣ dx≲εNε(N1/2q+N3/2q)(11)\mathbb{E}\sup_{f\in\mathcal{F}}\sum_{k=1}^{CN^4}\delta_k\int_{V_k}|f(x)|\,dx \lesssim_\varepsilon N^\varepsilon\left(N^{1/2}q+N^{3/2}\sqrt{q}\right) \tag*{(11)}

Remark 5. One could ask about the case when 1≤q≤N1 \le q \le N, however, this regime is not so interesting since in this case the random set VV is quite sparse and there exist f∈Ff \in\mathcal{F} for which ff is large on all of VV. In fact we have that when 1≤q≤N1 \le q \le N

Esup⁡f∈F∥f∥L1(V)∼qN.\mathbb{E}\sup_{f\in\mathcal{F}}\|f\|_{L^1(V)} \sim qN.

References

  1. [1]Jean Bourgain. Bounded orthogonal systems and the λ(p)-set problem. Acta Mathematica, 162(3–4):227–245, 1989.DOI
  2. [2]Jean Bourgain and Ciprian Demeter. The proof of the ℓ² decoupling conjecture. Annals of Mathematics, 182(1):351–389, 2015.arxiv.org/abs/1403.5335
  3. [3]Hannah Cairo and Ruixiang Zhang. Power loss for the mizohata–takeuchi conjecture on Cᵏ convex hypersurfaces, December 2025.arxiv.org/abs/2512.08064
  4. [4]Hannah Mira Cairo. A counterexample to the mizohata-takeuchi conjecture. arXiv preprint arXiv:2502.06137, 2025. Submitted on February 10, 2025.arxiv.org/abs/2502.06137
  5. [5]Emmanuel J. Candès and Yaniv Plan. A probabilistic and ripless theory of compressed sensing. 2010.arxiv.org/abs/1011.3854
  6. [6]Anthony Carbery, Marina Iliopoulou, and Hong Wang. Some sharp inequalities of Mizohata–Takeuchi-type. Revista Matemática Iberoamericana, 40(4):1387–1418, 2024.arxiv.org/abs/2302.11877
  7. [7]Yuqi Fu, Larry Guth, and Dominique Maldague. Sharp superlevel set estimates for small cap decouplings of the parabola. Revista Matemática Iberoamericana, 39(3):975–1004, 2023.arxiv.org/abs/2107.13139
  8. [8]Timothy Gowers. Inverse theorems in additive combinatorics. YouTube video (KSoM Channel). Video ID: Z3-v0apV-t0. YouTube lists it as ”Streamed 5 years ago” (accessed 2026-03-04).
  9. [9]Larry Guth. A restriction estimate using polynomial partitioning. Journal of the American Mathematical Society, 29(2):371–413, 2016.arxiv.org/abs/1407.1916
  10. [10]Larry Guth. Topics in analysis: Decoupling. Massachusetts Institute of Technology, 2017. Lecture notes for MIT 18.118, Fall 2017.
  11. [11]Larry Guth. Decoupling estimates in fourier analysis, 2022. Based on ICM 2022 survey lecture.arxiv.org/abs/2207.00652
  12. [12]Larry Guth. An enemy scenario in restriction theory. https://www.youtube.com/watch?v=x–DET83UJfG, 2022. Accessed: 2025-05-21.
  13. [13]Larry Guth. Sharp examples for decoupling and related questions. YouTube video, 2023. https://www.youtube.com/watch?v=h_pskQV–pU4.
  14. [14]Larry Guth, Jonathan Hickman, and Marina Iliopoulou. Sharp estimates for oscillatory integral operators via polynomial partitioning. Acta Mathematica, 223(2):251–376, 2019.arxiv.org/abs/1710.10349
  15. [15]Larry Guth, Dominique Maldague, and Hong Wang. Improved decoupling for the parabola, 2020.arxiv.org/abs/2009.07953
  16. [16]Alex Iosevich and Azita Mayeli. Uncertainty principles, restriction, Bourgain’s Λq theorem, and signal recovery. Applied and Computational Harmonic Analysis, 76(1):101734, 2024.
  17. [17]G. Mockenhaupt and W. Schlag. On the Hardy–Littlewood majorant problem for random sets. Journal of Functional Analysis, 256(4):1189–1237, 2009.
  18. [18]Siddharth Mulherkar. Random constructions for sharp estimates of mizohata–takeuchi type, June 2025.arxiv.org/abs/2506.05624
  19. [19]Michel Talagrand. Constructions of majorizing measures, bernoulli processes and cotype. Geometric and Functional Analysis, 4(6):660–717, 1994.arxiv.org/abs/math/9406216
  20. [20]Michel Talagrand. Upper and Lower Bounds for Stochastic Processes: Modern Methods and Classical Problems, volume 60 of Ergebnisse der Mathematik und ihrer Grenzgebiete. 3. Folge / A Series of Modern Surveys in Mathematics. Springer, 2014.DOI
  21. [21]Terence Tao. The uniform uncertainty principle and compressed sensing. 2008.
  22. [22]Ramon van Handel. Probability in high dimension. Lecture notes for APC 550, Princeton University, 2016.
  23. [23]Roman Vershynin. High-Dimensional Probability: An Introduction with Applications in Data Science, volume 47 of Cambridge Series in Statistical and Probabilistic Mathematics. Cambridge University Press, Cambridge, 2018.
  24. [24]Thomas H. Wolff. Lectures on Harmonic Analysis, volume 29 of University Lecture Series. American Mathematical Society, Providence, RI, 2003.DOI

Paper details

vlnc:W_2zvyv22u
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

Not recorded for this paper

View statement
Looking for related papers…
Contents