Damping oscillatory Integrals of convex analytic functions

Sanghyuk Lee; Sewook Oh

REVIEW 3 major objections 3 minor 46 references

Damping oscillatory Integrals of convex analytic functions

T0 review · 3 major / 3 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read For convex analytic hypersurfaces in dimensions 2–4, damping the surface measure by the square root of the Gaussian curvature yields the nondegenerate decay rate for the Fourier transform, up to a logarithm in dimension 4.

desk verdict Closes the convex-analytic damping problem in d=2,3 and misses d=4 only by a log; the proof is convincing, with two genuinely sketched cases that should be written out. read the letter →

arxiv 2505.15492 v2 pith:KZSU2JGA submitted 2025-05-21 math.CA

Sanghyuk Lee , Sewook Oh This is my paper · ORCID

classification math.CA MSC 42B2042B10

keywords oscillatoryintegralsdampingfactorGaussiancurvatureconvexhypersurfacesoffinitetypeFouriertransformdecayaffinerestrictionestimatesmaximaloperatorso-minimality

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 on a compact convex analytic hypersurface of finite type in dimensions two, three, and four, multiplying the surface measure by the square root of the Gaussian curvature (with a complex power $1/2+it$) restores the decay of the Fourier transform to essentially the rate one would get from a fully curved surface. Concretely, the Fourier transform decays like $C(1+|t|)^3|\xi|^{-d/2}$ for $d=2,3$ and like $C(1+|t|)^3(\log|\xi|)|\xi|^{-2}$ for $d=4$. This matters because earlier examples show the same decay is impossible in dimensions five and higher even for convex analytic surfaces, so the cases $2\le d\le 4$ were the remaining open range and the square-root curvature is the natural optimal damping order. The paper also shows that these Fourier decay estimates imply sharp endpoint estimates for the restriction, convolution, and maximal operators associated to the surface with the corresponding optimal damping factors.

What carries the argument

Two ingredients carry the proof. The first is a modification of the stationary set method for oscillatory integrals with a weight: the integral $\int e^{i(L\Phi_1+\tau\log\Phi_2)}a\,dx$ is bounded by an integral over dyadic level sets of $\Phi_2$ of $\sup_{\beta_1} S(y,\beta_1,\beta_2,L,\tau')$, where $S$ measures the mass of the amplitude on the set where $L\Phi_1\in[\beta_1,\beta_1+1]$ and $\Phi_2\in[\beta_2,\tau'\beta_2]$. The key estimate (3.4), which controls the oscillatory integral in $\beta_1$ by this supremum, uses the fact that $S$, as a function of $\beta_1$, changes monotonicity at most $N$ times with $N$ independent of the external parameters; that fact is imported from the o-minimal structure $\mathbb{R}_{an,exp}$ and the constructible-function integration theorem. The second ingredient is an affine normalization of the finite-type convex phase: after locating the critical point and rescaling via a convex-body normalization lemma so that the sublevel set $\{\Phi_v<h\}$ is comparable to the unit ball, the phase has uniform derivative bounds and its gradient is bounded below on the unit annulus, which permits the dyadic decomposition in $h$ and reduces the main estimate to the uniform bound in Proposition 4.8.

What would settle it

A concrete way to test the claim is to take a simple convex analytic finite-type phase such as $\varphi(x_1,x_2)=x_1^4+x_2^4+x_1^2x_2^2$ in dimension three, compute the function $S$ from Section 3.3 for large $L$ and $\tau'$, and check whether the number of monotonicity changes of $\beta_1\mapsto S(\beta_1,\beta_2,L,\tau')$ stays bounded independently of $\beta_2,L,\tau'$; if it grows, the stationary-set step fails. Alternatively, a direct numerical computation of $|(\kappa^{1/2}\sigma)^\wedge(\xi)|$ for a radial convex analytic surface such as $\varphi(x)=\frac14(|x|-1)^4+|x|$ in $\mathbb{R}^3$ could test whether the decay reaches $|\xi|^{-3/2}$; any sequence of frequencies where the decay is slower would disprove Theorem 1.1.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1.1: for $2\le d\le 4$, a convex analytic phase $\varphi$ of finite type, and $z=1/2+it$, the Fourier transform of the measure $\sigma_z$ with density $H_\varphi^{1/2+it}\psi$ obeys $|\widehat{\sigma_z}(\xi)|\le C(1+|t|)^3|\xi|^{-d/2}$ for $d=2,3$ and $C(1+|t|)^3(\log|\xi|)|\xi|^{-2}$ for $d=4$ when $|\xi|\ge 2$. Since the Hessian determinant $H_\varphi$ is comparable to the Gaussian curvature up to a smooth nonzero factor, this says that a half-power damping by curvature is enough to recover the nondegenerate decay exponent $-d/2$ in dimensions two and three, and the same exponent up to a logarithm in dimension four. The significance is that this damping order is the one conjecturally necessary for the optimal restriction, convolution, and maximal estimates, and the paper derives those consequences by analytic interpolation. The result does not extend to $d\ge 5$, where an earlier counterexample shows such decay can fail for convex analytic surfaces.

Load-bearing premise

The load-bearing premise is that the auxiliary function $S(y,\beta_1,\beta_2,L,\tau')$—the integrated measure of the level sets of the phase—switches between increasing and decreasing at most $N$ times in $\beta_1$, with $N$ independent of $y,\beta_2,L,\tau'$; the paper relies on an external theorem for this rather than proving it directly.

Editorial extensions

If this is right

For $d=2,3$, the $L^2$ restriction estimate with the affine surface measure $\kappa^{1/(d+2)}\sigma$ holds at the sharp endpoint exponent $p_\circ(d)=2(d+2)/(d+4)$; for $d=4$ it holds for $1\le p<p_\circ(d)$.
The convolution operator with damping exponent $1/(d+2)$ is $L^p\to L^q$ bounded on the full optimal triangle $T$ for $d=2,3$, and on the interior of $T$ plus the diagonal for $d=4$.
The maximal operator $M_{1/(d+1)}$ is bounded on $L^p$ exactly for $p>(d+1)/d$ when $2\le d\le 4$, and the damping order $1/(d+1)$ is optimal, so no smaller power can give the full range.
Because the constant $C(1+|t|)^3$ grows polynomially in $|t|$, the estimate is strong enough for analytic interpolation, which is why the three operator-theoretic corollaries follow from the single Fourier decay estimate.
Because the corresponding estimate is known to fail for $d\ge5$ even among convex analytic surfaces, the range $2\le d\le4$ is an essentially complete answer to the square-root damping question.

Reading between the lines

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

The analyticity assumption enters through definability in an o-minimal structure; if the same monotonicity-count bound held for smooth finite-type convex phases, the theorem would probably extend to them, so locating a smooth convex phase where $S$ changes monotonicity more than $N$ times would show analyticity is not merely technical.
The logarithmic factor in $d=4$ may be a removable endpoint artifact or may be genuine; comparing the sharp model in Remark 5.3 with higher-order radial examples could decide whether the $\log|\xi|$ reflects a true obstruction.
A direct analytic proof of the monotonicity bound for the specific weight $H_\varphi^{1/2+it}$ would replace the model-theoretic input with an explicit, quantitative argument and could yield effective constants, which matters if the estimate is used for dispersive or PDE applications.
The same weighted stationary-set machinery should apply to other damping exponents $\mathrm{Re}\,z$ close to $1/2$, producing a family of interpolating estimates that could refine the currently sharp-but-logarithmic $d=4$ case.

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.

Desk Editor's Note

read the letter

Lee and Oh settle the Cowling–Disney–Mauceri–Müller question for convex analytic hypersurfaces in dimensions two and three, and in four dimensions they are off by only a logarithm. That is the headline. The novelty is real: earlier work either required smoothness of the damping factor (Cowling et al.) or restricted to special surface classes (homogeneous, mixed homogeneous, radial). Here the square-root Hessian weight is handled for general convex analytic finite-type phases, with polynomial growth in |t|.

The paper does well the two things that matter most. The dyadic decomposition via John's lemma and the normalization of rescaled phases are executed carefully, and the key lemmas—uniform derivative bounds, the Glaeser-type estimate for derivatives of sqrt(H), the lower gradient bound on the dyadic shell—are proved. The adaptation of the stationary set method to a weight that is not smooth, via definability in R_an and integration closure in R_an,exp, is natural and, as far as I can see, correct. Proposition 3.6 (Cluckers–Miller) is a heavy external input, but it is a published theorem and the paper verifies its hypotheses honestly.

The soft spots are proportionate. The cases in Remark 5.5 and Section 5.3, where the gradient is bounded away from zero, are described as routine and many details are omitted. The main case would already make a strong paper; these two cases are needed for the full theorem, so the authors should write them out in the final version. I do not see a load-bearing flaw. The uniformity of the monotonicity-change bound in Theorem 3.7 is exactly the Basu–Guo–Zhang–Zorin-Kranich argument; the paper states the needed definability and cites the relevant proposition, though it would help to spell out the parameter uniformity once since the proof leans heavily on it.

Who is this for? Harmonic analysts working on oscillatory integrals, restriction theory, and convolution or maximal estimates for flat hypersurfaces. It deserves a serious referee and, modulo writing out the two sketched cases, publication. I would cite the d=2,3 result.

Referee Report

3 major / 3 minor

Summary. The paper proves optimal decay estimates for the Fourier transform of curvature-damped surface-carried measures on compact convex analytic hypersurfaces of finite type in dimensions 2, 3, and 4. Specifically, for a smooth measure σ on a graph H over a convex analytic phase φ of finite type and for z = 1/2 + i t, it establishes |(κ^{1/2+it}σ)^∧(ξ)| ≤ C(1+|t|)^3 |ξ|^{-d/2} for d = 2,3 and ≤ C(1+|t|)^3 log|ξ| |ξ|^{-2} for d = 4. The proof combines a dyadic decomposition of the phase level sets with a weighted version of the stationary set method of Basu–Guo–Zhang–Zorin-Kranich, using o-minimality to obtain uniform bounds on the number of monotonicity intervals of integrated counting functions. The paper also derives applications to L^2 restriction with affine surface measure, convolution estimates, and maximal operators with optimal damping orders.

Significance. If the main theorem is correct, it settles the remaining open cases 2 ≤ d ≤ 4 of the problem of optimal damping for convex analytic hypersurfaces, showing that the square-root curvature damping factor recovers the nondegenerate decay rate (up to a logarithm in d = 4) despite the nonsmoothness of κ^{1/2} on the flat set. This is a substantial advance over earlier results that required smoother damping factors. The adaptation of the stationary set method to amplitudes involving κ^{1/2} and phases with a logarithmic term is a genuinely new technical contribution. The paper is largely self-contained after invoking standard external tools (o-minimality of R_an,exp and the Cluckers–Miller integration theorem), and the auxiliary lemmas are proved in detail. The consequences for restriction, convolution, and maximal estimates are natural and are derived carefully, with the exception of a few presentation errors noted below.

major comments (3)

[§3.3, proof of Theorem 3.7, Eq. (3.3)] The change of variables in the proof of (3.3) is not correct as written. The initial display writes the β2 integral over the interval [-τ/|τ|, 0], so for τ > 0 this is [-1, 0]. But after the change of variables the integration limits become positive numbers Φ2 e^{-1/|τ|} and Φ2, and β2 appears inside a logarithm, so β2 must be positive. A logarithm of a negative number is involved in the displayed derivation, which is not meaningful. The intended identity is presumably of the form e^{iτ log Φ2} = C_τ ∫_0^∞ β2^{iτ-1} 1_{Φ2 ∈ [β2, e^{1/|τ|} β2]} dβ2, with β2 ranging over positive values. Since Theorem 3.7 is the core estimate used throughout Sections 5.1 and 5.2, this proof must be rewritten clearly; the final inequality (3.2) may still be correct, but the current text does not allow the reader to verify it.
[§5.2, use of (5.9) in the κ = 1 case] The estimate (5.9) is stated and proved only for the set {x ∈ supp A_{ℓ,0} : λhΦ_h(x) ∈ [β, β+1]}. In the proof of Proposition 4.9 for κ = 1, this estimate is used with the support of A_{ℓ,1} instead, without comment. The proof of (5.9) actually works for any amplitude whose support is contained in the set where |∂_ℓ Φ_h| ≥ c (which holds for both A_{ℓ,0} and A_{ℓ,1}), but the paper should state the general form and indicate that it applies to A_{ℓ,1}. This is a local gap, but it is load-bearing for the κ = 1 estimate (4.25).
[§4, proof of Theorem 2.2, Eq. (4.14)] The dyadic decomposition in (4.14) sums over all h ∈ D, but the subsequent estimate in the proof of Theorem 2.2 only sums over h < h0. The contribution of h ≥ h0 is not discussed. Since Φ_v is bounded on the support of ψ(x+ω_v), the number of dyadic scales h with h0 ≤ h ≤ ‖Φ_v‖_∞ is finite and the corresponding integrals I_h are O(1) by the trivial bound. This finite contribution can be absorbed into the final constant, but the paper should mention this to make the reduction from (4.14) to the h < h0 sum complete.

minor comments (3)

[§4, proof of Theorem 2.2, display after (4.17)] The integral displayed there, ∫_{2B_φ}^{1/λ} s^{d/2-1}(λs)^2 ds, has the integration limits reversed (since 2B_φ > 1/λ for large λ) and the integrand does not match the dyadic sum; the intended integral is λ^{-2} ∫_{1/λ}^{C} s^{d/2-3} ds, which yields the stated rates for d = 2,3,4.
[§A.1, proof of Corollary 1.3] The sentence 'it suffices to show that T is bounded from L^{d+2} to L^{(d+2)/(d+1)}' has the spaces reversed relative to (A.1), which states the operator maps L^{(d+2)/(d+1)} to L^{d+2}.
[§5.1, proof of (5.9)] The paper invokes Proposition 3.4 to conclude that S_{x'} is a union of at most N points and intervals, but Proposition 3.4 is stated for the number of monotonicity changes of a definable function. The uniform bound on the number of connected components of a definable family of subsets of R follows from o-minimality directly, and this should be stated explicitly for clarity.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: Theorem 1.1 is proved by a self-contained reduction to external o-minimality and convexity results, and the target estimate is never assumed.

full rationale

The derivation chain is not circular. The target estimate (1.2) is never used as an assumption; the proof reduces Theorem 1.1 to Theorem 2.1, then to Proposition 4.8 via a dyadic decomposition, and finally proves Proposition 4.8 using the stationary set method and the external o-minimality results Propositions 3.4 and 3.6. The paper explicitly imports these results from Basu–Guo–Zhang–Zorin-Kranich [1], Cluckers–Miller [8], and van den Dries–Macintyre–Marker [14]; they are independent mathematical theorems, not conclusions of the present paper. The only self-citation, [28] (Lee–Oh), does not appear in the body text and is therefore not load-bearing. The damping estimate is not obtained by fitting parameters or renaming a known quantity; the constants are uniform and the proof gives explicit reductions. The reliance on deep model-theoretic inputs such as o-minimality and constructible functions is a legitimate external dependency, and any failure there would be a correctness gap rather than circularity. No step equates a prediction with an input by construction, so the circularity score is 0.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

No data-fitted parameters exist in this proof; constants such as k, m, M, c_phi, C_phi, and h_0 are existential and depend on the fixed phase phi. The proof introduces no invented entities. It relies on standard model-theoretic results (o-minimality of R_an,exp, constructible-function integration), convex-geometric facts (John's lemma, finite-type convexity estimates), and the specific analytic finite-type assumption on phi.

assumptions (6)

standard math The structure R_an,exp is o-minimal (van den Dries, Macintyre, Marker).
Invoked in Section 3.2 (Theorem 3.5) to ensure definable functions change monotonicity boundedly many times.
standard math Constructible functions are stable under integration (Cluckers-Miller), so I_f(y) is definable in R_an,exp for R_an-definable f.
Proposition 3.6 is the bridge that lets the stationary set method handle R_an-definable phases after integration.
standard math A definable function h(t,y) changes monotonicity at most N times in t, with N independent of y (o-minimal monotonicity theorem).
Proposition 3.4, used to prove (3.4) in Theorem 3.7, which underlies the kappa=0 and kappa=1 estimates.
standard math John's lemma: for any compact convex set Omega with interior, there is an affine T with B_1 subset T^{-1}Omega subset B_d.
Lemma 4.3, used in the normalization argument of Section 4.1 to produce the rescaled phase Phi_h^v.
domain assumption Finite-type convex functions satisfy the derivative growth and oscillation bounds (4.1)-(4.2) and Lemma 4.1, from Bruna-Nagel-Wainger and Cowling et al.
This is the structural input that seeds all uniform bounds in Section 4.
domain assumption The phase phi is analytic, convex, finite-type on [-2,2]^d.
This is the class of hypersurfaces for which the theorem is stated; outside this class the estimate is known to fail for d>=5.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Damping oscillatory Integrals of convex analytic functions." pith.science (2026). https://pith.science/paper/KZSU2JGA

@misc{pith2026250515492,
  author       = {Pith},
  title        = {Pith review of: Damping oscillatory Integrals of convex analytic functions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KZSU2JGA}},
  note         = {Machine review of arXiv:2505.15492}
}

abstract

Let $H\subset \R^{d+1}$ be a compact, convex, analytic hypersurface of finite type with a smooth measure $\sigma $ on $H$. Let $\kappa$ denote the Gaussian curvature on $H$. We consider the oscillatory integral $(\kappa^{1/2} \sigma)^\wedge$ with the damping factor $\kappa^{1/2}$ and prove the optimal decay estimate \[ |(\kappa^{1/2} \sigma )^\wedge(\xi)|\le C|\xi|^{-d/2}\] for $d=2,3,$ and with an extra logarithmic factor for $d=4$. Our result provides an essentially complete answer, since such decay estimates generally fail for $d \ge 5$, even for convex analytic hypersurfaces, as shown by Cowling--Disney--Mauceri--M\"uller. Furthermore, we prove the same estimates for $(\kappa^{1/2+it} \sigma )^\wedge$ with $C$ growing polynomially in $|t|$. As consequences, we obtain the best possible estimates for the convolution, maximal, and adjoint restriction operators associated with $H$, incorporating the mitigating factors of optimal orders. In particular, for $d=2, 3$, we prove the $L^2$--$L^{2(d+2)/(d+4)}$ restriction estimate with respect to the affine surface measure $\kappa^{1/(d+2)} \sigma$. This work was inspired by the stationary set method due to Basu--Guo--Zhang--Zorin-Kranich.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 46 canonical work pages

[1]

S. Basu, S. Guo, R. Zhang, P. Zorin-Kranich,A stationary set method for estimating oscilla- tory integrals, J. Eur. Math. Soc. (2024), published online first, DOI 10.4171/JEMS/1506

work page doi:10.4171/jems/1506 2024
[2]

N. Bez, S. Lee, S. Nakamura,Strichartz estimates for orthonormal families of initial data and weighted oscillatory integral estimates, Forum Math. Sigma9(2021), Paper No. e1, 52 pp

work page 2021
[3]

Bourgain,Averages in the plane over convex curves and maximal operators, J

J. Bourgain,Averages in the plane over convex curves and maximal operators, J. Analyse Math.47(1986), 69–85

work page 1986
[4]

Bruna, A

J. Bruna, A. Nagel, S. Wainger,Convex hypersurfaces and Fourier transforms, Ann. of Math. 127(1988), 333–365

work page 1988
[5]

Carbery, M

A. Carbery, M. Christ, J. Wright,Multidimensional van der Corput and sublevel set estimates, J. Amer. Math. Soc.12(1999), 981–1015

work page 1999
[6]

Carbery, C

A. Carbery, C. E. Kenig, S. N. Ziesler,Restriction for homogeneous polynomial surfaces in R3, Trans. Amer. Math. Soc.365(2012), 2367–2407

work page 2012
[7]

Carbery, S

A. Carbery, S. N. Ziesler,Restriction and decay for flat hypersurfaces, Publ. Mat.46(2002), 405–434

work page 2002
[8]

Cluckers, D

R. Cluckers, D. J. Miller,Stability under integration of sums of products of real globally subanalytic functions and their logarithms, Duke Math. J.156(2011), 311–348

work page 2011

Show all 46 references

[9]

Cowling, G

M. Cowling, G. Mauceri,Inequalities for some maximal functions II, Trans. Amer. Math. Soc.296(1986), 341–365

1986
[10]

,Oscillatory integrals and Fourier transforms of surface carried measures, Trans. Amer. Math. Soc.304(1987), 53–68

1987
[11]

Cowling, S

M. Cowling, S. Disney, G. Mauceri, D. M¨ uller,Damping oscillatory integrals, Invent. Math. 101(1990), 237–260

1990
[12]

Dieudonn´ e,Sur un th´ eor` eme de Glaeser, Journal d’Analyse Math´ ematique,23(1970), 85–88

J. Dieudonn´ e,Sur un th´ eor` eme de Glaeser, Journal d’Analyse Math´ ematique,23(1970), 85–88

1970
[13]

van den Dries,Tame topology and o-minimal structures,London Mathematical Society Lecture note Series, 248, Cambridge University Press, 1998

L. van den Dries,Tame topology and o-minimal structures,London Mathematical Society Lecture note Series, 248, Cambridge University Press, 1998

1998
[14]

van den Dries, A

L. van den Dries, A. Macintyre, D. Marker,The elementary theory of restricted analytic fields with exponentiation, Ann. of Math.140(1994), 183–205

1994
[15]

Glaeser,Racine carr´ ee d’une fonction diff´ erentiable, Ann

G. Glaeser,Racine carr´ ee d’une fonction diff´ erentiable, Ann. Inst. Fourier (Grenoble)13 (1963), 203–207

1963
[16]

John Green,Uniform Oscillatory Integral estimates for Convex Phases via Sublevel Set es- timates, arXiv:2111.05395

arXiv
[17]

Greenblatt,Convexity, Fourier transforms, and lattice point discrepancy, arXiv: 2402.16636

M. Greenblatt,Convexity, Fourier transforms, and lattice point discrepancy, arXiv: 2402.16636

arXiv
[18]

Greenleaf,Principal curvature and harmonic analysis, Indiana Univ

A. Greenleaf,Principal curvature and harmonic analysis, Indiana Univ. Math. J.,30(1981), 519–537

1981
[19]

H¨ ormander,The Analysis of Linear Partial Differential Operators I: Distribution Theory and Fourier Analysis, 2nd ed., Springer-Verlag, Berlin (1990)

L. H¨ ormander,The Analysis of Linear Partial Differential Operators I: Distribution Theory and Fourier Analysis, 2nd ed., Springer-Verlag, Berlin (1990)

1990
[20]

Zametki78(2005), 833–852; translation in Math

I.A.Ikromov,Damped oscillatory integrals and maximal operators,Mat. Zametki78(2005), 833–852; translation in Math. Notes78(2005), 773–790. DAMPING OSCILLATORY INTEGRALS 29

2005
[21]

I. A. Ikromov, Sh. A. Muranov,On estimates for oscillatory integrals with a damping factor, Mat. Zametki104(2018), 200–215; translation in Math. Notes104(2018), 218–230

2018
[22]

Iosevich,Fourier transform,L 2 restriction theorem, and scaling, Boll

A. Iosevich,Fourier transform,L 2 restriction theorem, and scaling, Boll. Unione Mat. Ital. Sez. B Artic. Ric. Mat.8(1999), 383–387

1999
[23]

Iosevich, E

A. Iosevich, E. Sawyer,Oscillatory integrals and maximal averages over homogeneous sur- faces,Duke Math. J.82(1996), 103–141

1996
[24]

Math.132(1997), 46–119

,Maximal averages over surfaces,Adv. Math.132(1997), 46–119

1997
[25]

Iosevich, E

A. Iosevich, E. Sawyer, A. Seeger,On averaging operators associated with convex hypersur- faces of finite type, J. Anal. Math.79(1999), 159–187

1999
[26]

Kenig, G

C. Kenig, G. Ponce, L. Vega,Oscillatory integrals and regularity of dispersive equations, Indiana Univ. Math. J.40(1991), 33–69

1991
[27]

Kurdyka,On gradients of functions definable in o-minimal structures, Ann

K. Kurdyka,On gradients of functions definable in o-minimal structures, Ann. Inst. Fourier (Grenoble)48(1998), 769–783

1998
[28]

S. Lee, S. Oh,Uniform stationary phase estimate with limited smoothness, J. Geom. Anal. 34(2024), Paper No. 174

2024
[29]

Li,L 2 affine Fourier restriction theorems for smooth surfaces inR 3, to appear in Indiana University Mathematics Journal, arXiv:2210.15015v2

J. Li,L 2 affine Fourier restriction theorems for smooth surfaces inR 3, to appear in Indiana University Mathematics Journal, arXiv:2210.15015v2

arXiv
[30]

Littman,L p–Lq-estimates for singular integral operators arising from hyperbolic equa- tions, Partial differential equations (Proc

W. Littman,L p–Lq-estimates for singular integral operators arising from hyperbolic equa- tions, Partial differential equations (Proc. Sympos. Pure Math., Vol. 23, Univ. California, Berkeley, Calif., 1971), pp. 479–481, Amer. Math. Soc., Providence, RI, 1973

1971
[31]

Marletta,Maximal functions with mitigating factors in the plane, J

G. Marletta,Maximal functions with mitigating factors in the plane, J. London Math. Soc. 59(1999), 647–656

1999
[32]

Nagel, A

A. Nagel, A. Seeger, S. Wainger,Averages over convex hypersurfaces, Amer. J. Math.115(1993), 903–927

1993
[33]

Oberlin,Convolution estimates for some measures on curves, Proc

D. Oberlin,Convolution estimates for some measures on curves, Proc. Amer. Math. Soc.99 (1987), 56–60

1987
[34]

Scand.69(1991), 45–56

,Oscillatory integrals with polynomial phase, Math. Scand.69(1991), 45–56

1991
[35]

,Convolution with measures on hypersurfaces, Math. Proc. Cambridge Philos. Soc. 129(2000), 517–526

2000
[36]

,A uniform Fourier restriction theorem for surfaces inR 3,Proc. Amer. Math. Soc. 132(2004), 1195–1199

2004
[37]

Palle,Strichartz estimates for mixed homogeneous surfaces in three dimensions, Anal

L. Palle,Strichartz estimates for mixed homogeneous surfaces in three dimensions, Anal. PDE16(2023), 173–232

2023
[38]

Schlag,L p →L q stimates for the circular maximal function, Ph.D

W. Schlag,L p →L q stimates for the circular maximal function, Ph.D. Thesis. California Institute of Technology (1996)

1996
[39]

Schulz,Convex hypersurfaces of finite type and the asymptotics of their Fourier trans- forms, Indiana Univ

H. Schulz,Convex hypersurfaces of finite type and the asymptotics of their Fourier trans- forms, Indiana Univ. Math. J.40(1991), 1267–1275

1991
[40]

Shayya,An affine restriction estimate inR 3, Proc

B. Shayya,An affine restriction estimate inR 3, Proc. Amer. Math. Soc.135(2007), 1107– 1113

2007
[41]

C. D. Sogge, E. M. Stein,Averages of functions over hypersurfaces inR n, Invent. Math.82 (1985), 543–556

1985
[42]

E. M. Stein,Maximal functions: spherical means, Proc. Nat. Acad. Sci. U.S.A.73(1976), 2174–2175

1976
[43]

Stovall,Linear and bilinear restriction to certain rotationally symmetric hypersurfaces, Trans

B. Stovall,Linear and bilinear restriction to certain rotationally symmetric hypersurfaces, Trans. Amer. Math. Soc.369(2017), 4093–4117

2017
[44]

E. M. Stein,Harmonic Analysis; real-variable methods, orthogonality and oscillatory inte- grals, Princeton University Press (1993)

1993
[45]

E. M. Stein, G. Weiss,Introduction to Fourier Analysis in Euclidean Space, Princeton Uni- versity Press, Princeton 1971

1971
[46]

Tomas,A restriction theorem for the Fourier transform, Bull

P. Tomas,A restriction theorem for the Fourier transform, Bull. Amer. Mat. Soc.81(1975) 477–478. Department of Mathematical Sciences and RIM, Seoul National University, Seoul 08826, Republic of Korea Email address:shklee@snu.ac.kr 30 SANGHYUK LEE AND SEWOOK OH Center for Mathe...

1975

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

S. Basu, S. Guo, R. Zhang, P. Zorin-Kranich,A stationary set method for estimating oscilla- tory integrals, J. Eur. Math. Soc. (2024), published online first, DOI 10.4171/JEMS/1506

work page doi:10.4171/jems/1506 2024

[2] [2]

N. Bez, S. Lee, S. Nakamura,Strichartz estimates for orthonormal families of initial data and weighted oscillatory integral estimates, Forum Math. Sigma9(2021), Paper No. e1, 52 pp

work page 2021

[3] [3]

Bourgain,Averages in the plane over convex curves and maximal operators, J

J. Bourgain,Averages in the plane over convex curves and maximal operators, J. Analyse Math.47(1986), 69–85

work page 1986

[4] [4]

Bruna, A

J. Bruna, A. Nagel, S. Wainger,Convex hypersurfaces and Fourier transforms, Ann. of Math. 127(1988), 333–365

work page 1988

[5] [5]

Carbery, M

A. Carbery, M. Christ, J. Wright,Multidimensional van der Corput and sublevel set estimates, J. Amer. Math. Soc.12(1999), 981–1015

work page 1999

[6] [6]

Carbery, C

A. Carbery, C. E. Kenig, S. N. Ziesler,Restriction for homogeneous polynomial surfaces in R3, Trans. Amer. Math. Soc.365(2012), 2367–2407

work page 2012

[7] [7]

Carbery, S

A. Carbery, S. N. Ziesler,Restriction and decay for flat hypersurfaces, Publ. Mat.46(2002), 405–434

work page 2002

[8] [8]

Cluckers, D

R. Cluckers, D. J. Miller,Stability under integration of sums of products of real globally subanalytic functions and their logarithms, Duke Math. J.156(2011), 311–348

work page 2011