On Local Well-posedness of the Periodic Korteweg-de Vries Equation Below $H^{-\frac{1}{2}}(\mathbb{T})$

Ryan McConnell; Seungly Oh

REVIEW 3 major objections 6 minor 25 references

On Local Well-posedness of the Periodic Korteweg-de Vries Equation Below $H^{-\frac{1}{2}}(\mathbb{T})$

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read This paper proves that the periodic Korteweg-de Vries equation is locally well-posed in $H^s(\mathbb{T})$ for $-2/3 < s \le -1/2$ and small data, without using complete integrability.

desk verdict A new low-regularity range for periodic KdV (s > -2/3) with a promising data-dependent normal form method, but the final fixed-point contraction in Section 6 has a real gap: Gamma2 is written as a function of phi1 only while (19) still contains v, and the paper explicitly waves off the required substitution. read the letter →

arxiv 2411.15069 v1 pith:OXGVBQCC submitted 2024-11-22 math.AP

Ryan McConnell , Seungly Oh This is my paper · ORCID

classification math.AP MSC 35Q5337L5042B37

keywords periodicKdVequationlocalwell-posednessmodulation-restrictednormalformdata-dependentX^{sb}-typespacelow-regularitytransformationcancellationstructuresmalldata

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 local well-posedness of the periodic Korteweg-de Vries equation for real-valued mean-zero initial data in $H^s(\mathbb{T})$ whenever $-2/3 < s \le -1/2$ and the $H^s$ norm is sufficiently small. Previously, without using complete integrability, the best result was $s \ge -1/2$, where the standard bilinear $X^{s,b}$ estimate fails below that threshold. The proof builds a solution space $Y$ that depends on the initial datum and applies a modulation-restricted normal form to remove the resonant part of the nonlinearity, converting the difficult derivative loss into a time integral that can be controlled. If correct, this narrows the gap toward the integrable threshold $s \ge -1$ by a method that does not rely on inverse scattering.

What carries the argument

The mechanism is a modulation-restricted normal form $T^{\ell}(u,v) = T(\chi u, \chi v)$, where $\chi$ is a smooth Fourier cutoff that keeps only those bilinear interactions whose modulation $\langle \tau - n^3 \rangle$ is comparable to the resonant phase $|n(n_1+n_2)n_3|$; this removes the dangerous quadratic term while leaving bounded parts of the nonlinearity intact, preventing higher-order terms from acquiring unbounded modulation/frequency interactions. The argument runs in an initial-data-dependent $X^{s,b}$-type space $Y$ with weights $w_Y(n,L) = L^{1/2+}$ for low modulation and $L^{1/3+}\langle n\rangle^{1/3-}$ for high modulation, together with a smoother space $Z$, and uses the $L^6_{t,x}$ embedding supplied by $\ell^2$-decoupling. A cancellation lemma converts the resonant combination $\operatorname{Re}(r_n\overline{w_n}) + \tfrac{1}{2}|w_n|^2$ into an integral in time, so the derivative loss $2s-1$ is paid through the smoother $Z^*$ norm rather than through the failed bilinear estimate.

What would settle it

Compute the quadrilinear symbol bound claimed in Lemma 8, Case 3B(i), for frequencies $n_1 = N$, $n_3 = -N$, $N \gg |n_2|, |n_4|$: if the support of the difference $\chi(n_2,\tau_2;N) - \chi(n_2,\tau_2;-N)$ does not force $\langle \tau_2 - n_2^3 - \varphi_{n_2}\rangle \gtrsim N^2|n_2|$ when $\varphi_{n_2}$ is the data-dependent phase, then the claimed $\langle n_{\min}\rangle^{-1/2-}$ estimate fails and the contraction argument does not close.

Watch

Extended reading notes

Core claim

The central claim is Theorem 1: for real-valued mean-zero data $u_0 \in H^s(\mathbb{T})$ with $-2/3 < s \le -1/2$ and $\|u_0\|_{H^s}$ sufficiently small, there is a space $Y = Y(u_0)$ and a time $T = T(\|u_0\|_{H^s}) > 0$ such that $u_t + u_{xxx} = (u^2)_x$ has a unique solution $u \in C^0_t([0,T], H^s_x(\mathbb{T})) \cap Y$, and the data-to-solution map is continuous. The theorem is obtained without using the inverse-scattering structure of KdV, extending the previous non-integrable threshold $s \ge -1/2$ to $s > -2/3$. After rescaling the equation into $L^2$ by $u \mapsto \langle\nabla\rangle^{-s}u$, the proof writes $u = T^{\ell}(u,u) + v$ and uses a cutoff that restricts the normal form to modulations comparable to the resonant phase, so only an acceptable part of the quadratic nonlinearity is removed. The resonant remainder is absorbed into a modified linear propagator $W_t$ with phase $\varphi_n = \frac{2}{3}\langle n\rangle^{2s} n^{-1}|f_n|^2$, and the remaining nonlinearity is controlled in a pair of spaces $Y$ and $Z$.

Load-bearing premise

The load-bearing premise is that the formal substitution $u = T^{\ell}(u,u) + v$ can be inserted into the Duhamel nonlinearity after the time cutoff $\eta(t)$ is introduced, with the cutoff-induced extra terms either vanishing or falling into the already-controlled Case 3 estimates; Section 5.3 explicitly says this substitution detail is ignored for readability, and Section 6 asserts that the cutoff complications reduce to the spatial constraint of Case 3 without a full derivation.

Editorial extensions

If this is right

Local well-posedness for periodic KdV now holds for every $s > -2/3$ (small data), breaking the $s \ge -1/2$ barrier without using integrability.
The data-to-solution map is continuous, though by the known non-uniform-continuity result it cannot be uniformly continuous in this range.
The small-data restriction is part of the theorem; the fixed-point argument does not claim large-data well-posedness.
Uniqueness and continuous dependence hold inside the data-dependent space $Y$, so the solution map is well-defined from a small ball in $H^s$ into $C^0_t H^s_x \cap Y$.

Reading between the lines

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

The same modulation-restricted normal form could be tried on other quadratic dispersive equations on the torus whose resonant phase has a comparable factorization, potentially pushing their well-posedness thresholds below the $X^{s,b}$ bilinear endpoint.
Because the space $Y$ depends on the initial datum through $\varphi_n$, a natural test is whether a datum-independent version exists, or whether the small-data hypothesis can be removed by a different decomposition of the resonant dynamics.
The paper's own remarks in Sections 5.3 and 6 identify a concrete checkpoint: if the Littlewood-Paley reconstitution of $\eta T^{\ell}(\eta u, \eta u)$ cannot be closed, Theorem 1 would describe the time-cutoff equation but not the original KdV flow.

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

Two things to know. First, the claimed result is genuinely new: local well-posedness for periodic KdV for s > -2/3 without complete integrability, extending CKSTT's s = -1/2 by 1/6 derivative, via a data-dependent Bourgain space and modulation-restricted normal form adapted from NTT's mKdV work. Second, the proof as written has a real hole in the final contraction argument: Section 6 defines Gamma(phi1,phi2) = (Gamma1(phi1), Gamma2(phi1)), so Gamma2 depends only on phi1, while equation (19) contains v_n explicitly on the right-hand side. The paper says in Section 5.3 that the substitution u = h + v is needed and then ignores it "in favor of increased readability." That is not a minor detail; the Banach fixed-point step in Theorem 1 depends on making this substitution or redefining the map so that the second coordinate actually appears. The stress-test note is correct to flag this.

What the paper does well: the cancellation lemma (Lemma 4) is clean and the derivation of the data-dependent phase phi_n from the resonant term is natural, not fitted. Lemma 8's Case 3 quadrilinear reduction is the technical heart and the symbol estimates look plausible; the paper is honest about what is deferred. The citation pattern is appropriate: Oh's earlier work and NTT's mKdV paper are directly relevant, and self-citation there is not a problem.

Soft spots, in proportion: the fixed-point gap is load-bearing. The time-cutoff modifications in Section 6 are asserted rather than shown, and some frequency cases in Lemma 8 are omitted as "easier" or "omitted for brevity." Those are less concerning if the main gap closes. The small-data restriction is acceptable for this type of argument, and the failure to recover the conditional uniqueness range of NTT is not a flaw.

Who this is for: anyone working on low-regularity well-posedness of dispersive equations. The method is a useful template if the Section 6 gap is patched. It deserves a serious referee, but the referee should be asked to focus on the contraction map.

Recommendation: send to peer review, with the expectation of substantial revision. I would not cite the current version as a proven theorem, but I would read a revised version carefully.

Referee Report

3 major / 6 minor

Summary. The manuscript develops a modulation-restricted normal form method to prove local well-posedness of the periodic KdV equation in H^s(T) for s > -2/3, thereby extending the previously known threshold s >= -1/2 obtained via bilinear X^{s,b} estimates. After an L^2-based rescaling, the authors introduce a data-dependent phase φ_n and define associated Bourgain-type spaces Y and Z. They split the solution as u = T^ell(u,u) + v, derive a Duhamel equation (19) for v, and state a series of multilinear estimates (Lemmas 5-10) for the nonlinear terms. Theorem 1 then asserts existence, uniqueness, and continuity of the data-to-solution map for sufficiently small data in a data-dependent space Y, with the proof intended to follow from a Banach fixed-point argument in Section 6.

Significance. If the result is correct, it is significant: it would give the first local well-posedness result for periodic KdV below H^{-1/2} without using complete integrability, matching the mKdV analogue of Nakanishi, Takaoka, and Tsutsumi up to a small-data caveat. The modulation-restricted normal form and the spectral-phase-dependent Bourgain spaces are novel tools for KdV and are likely to be reusable. The paper contains a substantial amount of detailed symbolic and case-based analysis, and the data-dependent phase is derived from the resonant term rather than fitted to force the conclusion. However, as written, the central contraction argument has load-bearing gaps that prevent the theorem from being established.

major comments (3)

[Section 6, definition of Γ] The map Γ(φ1, φ2) is defined as (Γ1(φ1), Γ2(φ1)), with Γ2 depending only on φ1, but the right-hand side of equation (19) contains v_n explicitly, both in the integral term ∫_0^t Re(v_n N_R)(s)ds and in the terms of N_R listed in (18), such as R(u,u,u) - R(v,v,v). Since v is not replaced by u - h inside Γ2 as written, the displayed Γ2 is not a functional of φ1 only, and the subsequent contraction estimate comparing Γ2(φ1) with Γ2(ψ1) cannot produce the claimed factor ||φ2 - ψ2||_{Z1}. This gap undermines the Banach fixed-point step and hence the uniqueness and continuous-dependence conclusions in Theorem 1.
[Section 5.3, Remark 4] The proof of Lemma 8, Case 3, uses the decomposition u = h + v, and Remark 4 states that the full equation should contain M^{(1,2)}(u,u,u) + M^{(3)}(h+v,u,u) in place of M(u,u,u). However, this substitution is not implemented in the fixed-point equation of Section 6; Section 5.3 explicitly says 'We ignore this minor detail in favor of increased readability.' As a result, the estimates of Lemma 8 are not shown for the actual contraction map, and this is load-bearing because Lemma 8 is one of the principal estimates used to bound N_R in (30).
[Section 6, time cutoff] The passage from the original equation (1) to the time-cutoff equation (31) is asserted rather than proved. The text states that η ∈ H^8_t and that 'all of our spaces are bounded with respect to this cutoff', allowing the authors to 'disregard this minor technical difficulty.' Similarly, the compatibility of the cancellation in Lemma 4 with the cutoff is hand-waved by saying that a reconstitution of Littlewood-Paley blocks 'reduces us purely to the spatial constraint of Case 3.' Since the fixed-point argument is run on the cutoff equation, these assertions need to be demonstrated, especially because the data-dependent phase φ_n and the resonance removal are sensitive to the exact form of the nonlinearity.

minor comments (6)

[Section 2] The substitution u ↦ <∇>^{-s}u and the subsequent renaming s -> |s| is confusing; the reader must keep track that after the substitution, positive s in [1/2, 2/3) corresponds to negative regularity in Theorem 1. It would help to state this correspondence explicitly.
[Section 5.2] The functional }v}_X = min(}v}_\tilde{X}, }v}_{X^{2s-1+, -1/2+}}) is not a norm, since the minimum of two norms is not generally a norm, yet it is used in Lemma 6 and in (30) as though it were.
[Section 5.2, Lemma 6] The notation }K}_X is used for a functional applied to K, and the text acknowledges that X lacks the triangle inequality but proceeds to treat it as one; this requires clarification or a reformulation as a genuine norm.
[Section 5.3 heading] The heading contains a typo: 'nonlinearlity' should be 'nonlinearity'.
[Lemma 3] The passage from the ℓ2-decoupling estimate to the claimed embedding X^{0+,1/2+} ↦ L^6_{t,x} is not fully justified; the proof cites [25, Lemma 2.9] but does not show how the frequency-dependent phase φ_n fits into that lemma.
[Section 4, equation (19)] The term N_R is defined only by the list in (18); it would help readability to write out the full expression for N_R in a displayed equation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the data-dependent phase is computed from the resonant term, and the main estimates do not reduce to the theorem's assumptions.

full rationale

The paper's central new device is the initial-data-dependent phase φ_n = (2/3)(<n>^{2s}/n)|f_n|^2. This is not an adjustable parameter fitted to force the conclusion: Section 4 derives it from the resonant coefficient R_n = i(<n>^{2s}/n)|u_n|^2 u_n, and the phase is then used to define the modified linear evolution L_n = n^3 + φ_n. The choice is canonical for cancelling the resonant interaction, not an input that already contains the stated well-posedness result. The auxiliary space Y = Y(u0) depends on the initial data through φ_n, but the theorem's conclusion is stated in the fixed H^s topology, and continuous dependence in that topology is proved separately in Lemma 12; data-dependence of an auxiliary norm is not definitional circularity and is standard in the mKdV precedents [18, 24]. External ingredients, including Schippa's ℓ^2-decoupling theorem, the Ginibre–Tsutsumi–Velo lemma, and Bourgain's X^{s,b} framework, are independent support. The self-citations to [19], [15], and [20] are not load-bearing in a circular way: [19] supplies a direct algebraic formula for the resonant coefficient, while [15] and [20] provide background smoothing results. The manuscript does contain explicitly flagged technical gaps: Section 5.3 assumes 'the system of equations given by u = h+v' and states 'our fixed point equation will need to have this substitution reflected. We ignore this minor detail in favor of increased readability'; Section 6 asserts that cutoff-induced complications 'reduce us purely to the spatial constraint of Case 3' without a full derivation; and the displayed fixed-point map Γ(ϕ1,ϕ2) = (Γ1(ϕ1), Γ2(ϕ1)) does not visibly reflect the v-dependence of equation (19). These are rigor and completeness gaps, not circularity: they are unexecuted substitutions or omitted justifications, and repairing them would require additional estimates rather than making the theorem true by construction.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No free constants are fitted to data; the implicit constants in 'À' are absolute or depend on f and s. The only data-dependent quantity is the phase φ_n entering the norm, which is derived from the resonant term, not chosen to match outcomes. No new physical entities are postulated.

assumptions (5)

standard math Standard Fourier series and X^{s,b} functional framework for periodic dispersive equations
Used throughout Sections 2-5 to define spaces and norms.
standard math Schippa's ℓ2-decoupling estimate [21, Proposition 1] gives L^6_{t,x} embedding for the phase ξ^3+G(ξ)
Invoked in Lemma 3 to prove X^{0+,1/2+} ⊂ L^6_{t,x}.
standard math Lemma 2.1 of Ginibre, Tsutsumi, Velo [10] on inhomogeneous Besov-type estimates
Used in Lemma 6 to bound the integral term in the Duhamel formulation.
domain assumption Mean-zero and real-valued initial data u0 ∈ H^s, with frequencies n ≠ 0
The mean-zero condition is conserved and is assumed from the start; the real-valued condition is part of Theorem 1.
domain assumption Smallness of ||u0||_{H^s} ensures the data-dependent phase φ_n is a small perturbation of n^3 so that decoupling and C^2 phase properties hold
Used in Lemma 3 and the contraction argument; the theorem explicitly restricts to sufficiently small data.

how reviews work

0 comments

Cite this review

Pith. "Pith review of On Local Well-posedness of the Periodic Korteweg-de Vries Equation Below $H^{-\frac{1}{2}}(\mathbb{T})$." pith.science (2026). https://pith.science/paper/OXGVBQCC

@misc{pith2026241115069,
  author       = {Pith},
  title        = {Pith review of: On Local Well-posedness of the Periodic Korteweg-de Vries Equation Below $H^-\frac12(\mathbbT)$},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OXGVBQCC}},
  note         = {Machine review of arXiv:2411.15069}
}

abstract

We utilize a modulation restricted normal form approach to establish local well-posedness of the periodic Korteweg-de Vries equation in $H^s(\mathbb{T})$ for $s> -\frac23$. This work creates an analogue of the mKdV result by Nakanishi, Takaoka, and Tsutsumi for KdV, extending the currently best-known result of $s \geq -\frac12$ without utilizing the theory of complete integrability.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 24 canonical work pages

[1]

Babin, Alexei A

Anatoli V . Babin, Alexei A. Ilyin, and Edriss S. Titi. On t he regularization mechanism for the periodic Korteweg-de Vries equation. Comm. Pure Appl. Math. , 64(5):591–648, 2011

work page 2011
[2]

Sharp well-posedness and ill-posedness results for a quadratic non-linear Schr¨ odinger equation.J

Ioan Bejenaru and Terence Tao. Sharp well-posedness and ill-posedness results for a quadratic non-linear Schr¨ odinger equation.J. Funct. Anal., 233(1):228–259, 2006

2006
[3]

Bourgain

J. Bourgain. Fourier transform restriction phenomena f or certain lattice subsets and applications to nonlinear ev o- lution equations. I. Schr¨ odinger equations.Geom. Funct. Anal., 3(2):107–156, 1993

work page 1993
[4]

Bourgain

J. Bourgain. Fourier transform restriction phenomena f or certain lattice subsets and applications to nonlinear ev o- lution equations. II. The KdV-equation. Geom. Funct. Anal., 3(3):209–262, 1993

work page 1993
[5]

Asy mptotics, frequency modulation, and low regularity ill- posedness for canonical defocusing equations

Michael Christ, James Colliander, and Terrence Tao. Asy mptotics, frequency modulation, and low regularity ill- posedness for canonical defocusing equations. Amer . J. Math., 125(6):1235–1293, 2003

work page 2003
[6]

Colliander, M

J. Colliander, M. Keel, G. Stafﬁlani, H. Takaoka, and T. T ao. Sharp global well-posedness for KdV and modiﬁed KdV on R and T. J. Amer . Math. Soc., 16(3):705–749, 2003

work page 2003
[7]

Global smo othing for the periodic KdV evolution

Mehmet Burak Erdo˘ gan and Nikolaos Tzirakis. Global smo othing for the periodic KdV evolution. Int. Math. Res. Not. IMRN, (20):4589–4614, 2013

work page 2013
[8]

Burak Erdo˘ gan and Nikolaos Tzirakis

M. Burak Erdo˘ gan and Nikolaos Tzirakis. Dispersive partial differential equations, volume 86 of London Mathe- matical Society Student Texts. Cambridge University Press, Cambridge, 2016. Wellposedn ess and applications

work page 2016

Show all 25 references

[9]

Burak Erdo˘ gan, T

M. Burak Erdo˘ gan, T. Burak G¨ urel, and Nikolaos Tzirakis. Smoothing for the fractional Schr¨ odinger equation on the torus and the real line. Indiana Univ. Math. J., 68(2):369–392, 2019

2019
[10]

Ginibre, Y

J. Ginibre, Y . Tsutsumi, and G. V elo. On the Cauchy problem for the Zakharov system. J. Funct. Anal., 151(2):384– 436, 1997

1997
[11]

Kappeler and P

T. Kappeler and P . Topalov. Global wellposedness of KdV in H ´ 1pT, Rq. Duke Math. J., 135(2):327–360, 2006

2006
[12]

Low regularity well-posedness for the p eriodic Kawahara equation

Takamori Kato. Low regularity well-posedness for the p eriodic Kawahara equation. Differential Integral Equations, 25(11-12):1011–1036, 2012

2012
[13]

Cancellation proper ties and unconditional well-posedness for the ﬁfth order KdV type equations with periodic boundary condition

Takamori Kato and Kotaro Tsugawa. Cancellation proper ties and unconditional well-posedness for the ﬁfth order KdV type equations with periodic boundary condition. Partial Differ . Equ. Appl., 5(3):Paper No. 18, 55, 2024

2024
[14]

Kenig, Gustavo Ponce, and Luis V ega

Carlos E. Kenig, Gustavo Ponce, and Luis V ega. A bilinear estimate with applications to the KdV equation. J. Amer . Math. Soc., 9(2):573–603, 1996

1996
[15]

Well-posedness for the non-integrabl e periodic ﬁfth order KdV in Bourgain spaces

Ryan McConnell. Well-posedness for the non-integrabl e periodic ﬁfth order KdV in Bourgain spaces. Discrete Contin. Dyn. Syst. , 44(11):3507–3552, 2024

2024
[16]

Robert M. Miura. Korteweg-de Vries equation and genera lizations. I. A remarkable explicit nonlinear transforma- tion. J. Mathematical Phys., 9:1202–1204, 1968

1968
[17]

Sharp ill-posedness results for the KdV an d mKdV equations on the torus

Luc Molinet. Sharp ill-posedness results for the KdV an d mKdV equations on the torus. Adv. Math., 230(4-6):1895– 1930, 2012

1930
[18]

L ocal well-posedness in low regularity of the mKdV equa- tion with periodic boundary condition

Kenji Nakanishi, Hideo Takaoka, and Y oshio Tsutsumi. L ocal well-posedness in low regularity of the mKdV equa- tion with periodic boundary condition. Discrete Contin. Dyn. Syst. , 28(4):1635–1654, 2010

2010
[19]

Resonant phase-shift and global smoothing of the periodic Korteweg-de Vries equation in low regu- larity settings

Seungly Oh. Resonant phase-shift and global smoothing of the periodic Korteweg-de Vries equation in low regu- larity settings. Adv. Differential Equations, 18(7-8):633–662, 2013

2013
[20]

Stefanov

Seungly Oh and Atanas G. Stefanov. Smoothing and growth bound of periodic generalized Korteweg–De Vries equation. J. Hyperbolic Differ . Equ., 18(4):899–930, 2021

2021
[21]

On strichartz estimates from ℓ2-decoupling and applications

Robert Schippa. On strichartz estimates from ℓ2-decoupling and applications. In Willy D¨ orﬂer, Marlis Hochbruck, Dirk Hundertmark, Wolfgang Reichel, Andreas Rieder, Rolan d Schnaubelt, and Birgit Sch¨ orkhuber, editors,Math- ematics of W ave Phenomena, pages 279–289, Cham, 20...

2020
[22]

Normal forms and quadratic nonlinear Kle in-Gordon equations

Jalal Shatah. Normal forms and quadratic nonlinear Kle in-Gordon equations. Comm. Pure Appl. Math., 38(5):685– 696, 1985

1985
[23]

Nonlinear smoothing for the periodic di spersion generalized benjamin-ono equations with polyno- mial nonlinearity, 2024

Wangseok Shin. Nonlinear smoothing for the periodic di spersion generalized benjamin-ono equations with polyno- mial nonlinearity, 2024. 26 RY AN MCCONNELL, SEUNGL Y OH

2024
[24]

Well-posedness of t he Cauchy problem for the modiﬁed KdV equation with periodic boundary condition

Hideo Takaoka and Y oshio Tsutsumi. Well-posedness of t he Cauchy problem for the modiﬁed KdV equation with periodic boundary condition. Int. Math. Res. Not. , (56):3009–3040, 2004

2004
[25]

Nonlinear dispersive equations , volume 106 of CBMS Regional Conference Series in Mathematics

Terence Tao. Nonlinear dispersive equations , volume 106 of CBMS Regional Conference Series in Mathematics . Conference Board of the Mathematical Sciences, Washington , DC; by the American Mathematical Society, Provi- dence, RI, 2006. Local and global analysis

2006

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Babin, Alexei A

Anatoli V . Babin, Alexei A. Ilyin, and Edriss S. Titi. On t he regularization mechanism for the periodic Korteweg-de Vries equation. Comm. Pure Appl. Math. , 64(5):591–648, 2011

work page 2011

[2] [2]

Sharp well-posedness and ill-posedness results for a quadratic non-linear Schr¨ odinger equation.J

Ioan Bejenaru and Terence Tao. Sharp well-posedness and ill-posedness results for a quadratic non-linear Schr¨ odinger equation.J. Funct. Anal., 233(1):228–259, 2006

2006

[3] [3]

Bourgain

J. Bourgain. Fourier transform restriction phenomena f or certain lattice subsets and applications to nonlinear ev o- lution equations. I. Schr¨ odinger equations.Geom. Funct. Anal., 3(2):107–156, 1993

work page 1993

[4] [4]

Bourgain

J. Bourgain. Fourier transform restriction phenomena f or certain lattice subsets and applications to nonlinear ev o- lution equations. II. The KdV-equation. Geom. Funct. Anal., 3(3):209–262, 1993

work page 1993

[5] [5]

Asy mptotics, frequency modulation, and low regularity ill- posedness for canonical defocusing equations

Michael Christ, James Colliander, and Terrence Tao. Asy mptotics, frequency modulation, and low regularity ill- posedness for canonical defocusing equations. Amer . J. Math., 125(6):1235–1293, 2003

work page 2003

[6] [6]

Colliander, M

J. Colliander, M. Keel, G. Stafﬁlani, H. Takaoka, and T. T ao. Sharp global well-posedness for KdV and modiﬁed KdV on R and T. J. Amer . Math. Soc., 16(3):705–749, 2003

work page 2003

[7] [7]

Global smo othing for the periodic KdV evolution

Mehmet Burak Erdo˘ gan and Nikolaos Tzirakis. Global smo othing for the periodic KdV evolution. Int. Math. Res. Not. IMRN, (20):4589–4614, 2013

work page 2013

[8] [8]

Burak Erdo˘ gan and Nikolaos Tzirakis

M. Burak Erdo˘ gan and Nikolaos Tzirakis. Dispersive partial differential equations, volume 86 of London Mathe- matical Society Student Texts. Cambridge University Press, Cambridge, 2016. Wellposedn ess and applications

work page 2016