Generic Spectral Determination of Semiclassical Schr\"odinger Operators with $\mathbb Z_2$-Symmetry

Qiaoling Wei

Generic Spectral Determination of Semiclassical Schr\"odinger Operators with $\mathbb Z_2$-Symmetry

T0 review · 0 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 The low-energy spectrum generically determines a symmetric potential up to inversion.

desk verdict Solid, honest extension of Wang26: the B[4] layer generically kills the affine defect left by the first two QBNF layers, and the proof checks out. read the letter →

arxiv 2608.11111 v1 pith:JGCE6EST submitted 2026-08-11 math.DS

Qiaoling Wei This is my paper · ORCID

classification math.DS MSC 35P2081Q2035R3053D50

keywords semiclassicalinversespectralproblemquantumBirkhoffnormalformZ2-symmetryMoyalproductdeterminationgenericinjectivitySchrödingeroperatorTaylorseriesatawell

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 asks the quantum version of the drum problem: can the low-energy spectrum of a Schrödinger operator on the plane determine the potential? The author proves that, generically, the answer is yes for potentials with one reflection symmetry, up to the unavoidable reflection $V(x)\mapsto V(-x)$. Specifically, for a nondegenerate well at the origin with rationally independent harmonic frequencies, the first three layers of the quantum Birkhoff normal form---$B^{[0]}$, $B^{[2]}$, $B^{[4]}$---determine the complete Taylor series of the potential at the well, provided the ten-jet of the potential avoids a meager exceptional set. For real analytic potentials with a unique isolated well, the same conclusion follows from equality of the low-lying semiclassical spectra. The payoff is that a finite, explicitly checkable jet condition turns the first three quantum layers into a complete Taylor-series bookkeeper, so the generic symmetric well is spectrally unique up to spatial inversion.

What carries the argument

The quantum Birkhoff normal form (QBNF) is the unique resonant normal form obtained by a formal Moyal automorphism (a formal change of variables compatible with the noncommutative product of quantum observables); its layers $B^{[2r]}$ are the coefficients of $\hbar^{2r}$ in the normalized symbol, and 'resonant' means commuting with the harmonic-oscillator part. Three linked mechanisms carry the argument. First, the normal-form map is triangular in the Weyl grading (with $\deg x_j=\deg\xi_j=1$ and $\deg\hbar=2$): at each degree $2N$ the block operator $A_N$ that sends the new pair $(V_{2N-1},V_{2N})$ to the recorded $B^{[0]}$ and $B^{[2]}$ coefficients has exactly a one-dimensional kernel spanned by $(\Phi_N,\Psi_N)$, so the first two layers leave an affine line of possibilities parametrized by the edge coefficient $a_{1,2N-2}$. Second, the $\hbar^4$ layer enters through a transfer operator $T_V$; the scalar edge functions $d_N$ couple to that kernel, making each edge coefficient observable, while $B^{[4]}_8$ reduces the first ambiguous edge coefficient $a_{14}$ to at most two candidates and a compatibility determinant rejects the false branch. Third, global separation of the initial cubic pair is a finite-dimensional incidence statement: a polynomial map $I_{10}$ on the 32-dimensional ten-jet space has generic rank 30, and the additional $\hbar^4$ data $B^{[4]}_{12}$ cut the incidence set below the dimension of the jet space, with Lemma 6.5's surjectivity making the cut effective.

What would settle it

Compute, for a concrete pair of cubic coefficients with $(a_{30},a_{12})\ne \pm(a_{30}^*,a_{12}^*)$, the $3\times 3$ matrix of the map $G\mapsto \frac12\Pi(M_3(X^*)^2-M_3(X)^2)G$ from degree-ten potential polynomials to the resonant span $\{\Omega_1^2,\Omega_1\Omega_2,\Omega_2^2\}$. If the rank is ever less than three, Lemma 6.5 fails and the incidence-dimension argument of Proposition 6.7 loses its codimension; a direct counterexample to Theorem 1.1 would be two sign-compatible solutions of the compatibility system $P_{Y,Z}=0$ arising from one generic ten-jet.

Watch

Extended reading notes

Core claim

At the level of formal Taylor series, the central claim is a generic injectivity theorem for the quantum Birkhoff normal form map. For each sign $\sigma$ of the leading cubic coefficient $a_{30}$, there is a dense $G_\delta$ set of ten-jets such that any two $\mathbb{Z}_2$-symmetric formal potentials whose jets lie in that set and whose QBNF layers $B^{[0]}$, $B^{[2]}$, $B^{[4]}$ agree must be identical; without fixing the sign, they differ by at most the spatial inversion $\iota V(x)=V(-x)$. At the spectral level, for real analytic trapped potentials, equality of the semiclassical spectrum---or equality of low-energy fixed-window spectra for all sufficiently small $\hbar$---forces $W=V$ or $W=\iota V$. The reconstruction is effective: after the initial cubic pair is separated by a finite compatibility system, the remaining Taylor coefficients are recovered recursively from the three layers. Thus the first three QBNF layers generically encode the complete Taylor series at the well, and the only generic spectral ambiguity is the reflection symmetry inherent in the problem.

Load-bearing premise

The load-bearing premise is that genuinely different cubic terms always generate enough independent degree-twelve $\hbar^4$ data to rule out a coincidence; if the map $\frac12\Pi(M_3(X^*)^2-M_3(X)^2)$ failed to be surjective for some pair of cubic terms, the exceptional set in Proposition 6.7 could grow from a meager set to a positive-dimensional family.

Editorial extensions

If this is right

For a generic symmetric well, the complete Taylor series at the well and, for real analytic potentials, the potential itself are determined by the low-lying spectrum up to the reflection $x\mapsto -x$; no additional normalisation or transverse family is needed.
The exceptional potentials form a meager, nowhere dense set in the ten-jet space, so any non-uniqueness is a rare phenomenon rather than an open family of counterexamples.
The first two QBNF layers leave one scalar ambiguity at every degree, and the $\hbar^4$ layer resolves it; this explains why three layers suffice and where each layer's information resides.
Equality of the exact low-energy fixed-window spectra for all sufficiently small $\hbar$ gives the same conclusion as equality of the semiclassical spectrum, so the result applies to a finite energy window rather than only to an asymptotic family.

Reading between the lines

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

Beyond the paper: the filtered block structure with finite-dimensional kernels and a higher-layer coupling is a plausible template for other symmetry groups; one would expect fewer layers to suffice with more symmetry and more layers with none.
Beyond the paper: because the compatibility data are encoded in a finite polynomial system, the reconstruction is in principle algorithmic; the paper does not address numerical stability or the effect of finite-precision spectral data.
Beyond the paper: the theorem is generic, so it leaves open whether any ten-jet in the exceptional set actually gives rise to a distinct potential with identical first three layers; constructing or ruling out such an example would clarify whether the genericity hypothesis is intrinsic or only a proof device.

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

Bottom line: the result is real and the proof is sound. The genuinely new step is using the B[4] layer to kill the one-dimensional affine defect left by B[0] and B[2] at every degree, plus the finite-jet separation of the initial cubic pair. The paper is honest about what is new: Theorem 3.5 is credited to Wang26, and the contribution is the structural reformulation and the B[4] argument built around it.

What it does well: the algebraic filtration is organized cleanly. Proposition 2.10 gives a triangular block operator, Theorem 3.4 isolates the affine defect exactly, and Section 4 converts each defect into a scalar observable d_N. I checked the Fischer-adjoint step in Lemma 6.5/6.6: the surjectivity claim is correct, and the dimension count in Proposition 6.7 follows. The main theorems are stated carefully, with the sign/inversion ambiguity handled properly. The generic conditions depend only on the ten-jet, which is the right level of explicitness for this method.

Soft spots, in proportion: the result is generic rather than explicit. The dense G_delta sets are defined by nonvanishing of rational functions plus a Sard/measure-zero exclusion, so you cannot exhibit a concrete admissible jet. That is normal for this kind of argument, not a flaw. Theorem 1.4 relies on the standard CdV09 spectral-to-QBNF equivalence, which is cited rather than reproved; fine, but the spectral conclusion inherits a substantial external theorem. The paper is long and sometimes terse, and the topology appendix is dense. Also, only the three-dimensional resonant projection of B[4]_12 is used in the separation step; that turns out to be enough.

I read the flagged Lemma 6.5/6.6 carefully. The stress-test concern does not land: T = P(∂) has injective Fischer adjoint, hence is surjective; E2 is an isomorphism; the tangent vector (G, G) is legitimate. The fragile step holds.

Overall: a meaningful advance within an established program, not a paradigm shift. It deserves a serious referee. I would send it out, ideally to someone who knows semiclassical normal forms. I would cite it if I worked in inverse spectral theory.

Referee Report

0 major / 6 minor

Summary. The paper studies the semiclassical inverse spectral problem for P_ℏ = -ℏ²Δ/2 + V on R², with V even in x₂, nonresonant frequencies, and a nondegenerate well at the origin. It proves a generic injectivity theorem at the level of the quantum Birkhoff normal form: for each sign σ there is a dense Gδ subset of the ten-jet space such that if V and W lie in the fixed-sign formal potential space and share the first three QBNF layers B[0], B[2], B[4], then W = V. Equivalently, for such V the first three layers determine the complete Taylor series at the well. A sign-free version gives W = V or W = V(-x); real analytic and trapped versions (Theorems 1.3 and 1.4) pass from QBNF data to low-lying semiclassical spectra via the cited Colin de Verdière spectral-to-QBNF equivalence. The proof is organized as a filtered homological-equation calculation: the first two layers leave exactly one affine defect per degree, whose parameters are shown to be detected generically by the ℏ⁴ layer; low-order compatibility then fixes a14, and a finite-dimensional Sard/incidence argument separates the cubic pair up to sign.

Significance. If correct, this is a substantial advance in the bottom-of-the-well inverse spectral problem in dimension two: prior results required additional data or symmetry beyond a single reflection to recover the Taylor series from one or two QBNF layers, whereas here three layers generically encode the full jet. The proof is unusually concrete: the block operators A_N, the explicit transfer operator T_V, and the derivative computations (e.g. ∂d_N/∂a50 in Proposition 4.11) are directly checkable. I specifically checked the most fragile point identified in the stress test, the surjectivity of 1/2Π(M_3(X*)² - M_3(X)²) in Lemma 6.5 and the resulting submersion Θ in Lemma 6.6; the Fischer-adjoint argument and the dimension count dim R = 34, dim Z ≤ 31 are sound. The paper is also honest about its single external input, the spectral-to-QBNF equivalence cited from [CdV09], and it does not rely on fitted parameters or post-hoc selection of the generic set. If the results are accepted, the paper likely becomes a reference for generic spectral rigidity with Z₂-symmetry.

minor comments (6)

[Corollary 4.10] The statement that z_N(V) depends only on the (2N−2)-jet and B[4](V) omits the dependence of the base point (\hat V_{2N−1}, \hat V_{2N}) on B[2]_{2N} and B[0]_{2N}; since those layers are part of the QBNF data, this does not affect the argument, but the wording should be corrected.
[Proposition 4.11 and Lemma 5.10] The displayed identities d_N(V + εx_1^5) = d_N(V) + εm_N(V) and d_4(W_ε) = d_4(V^♯) + εm_4 are written as exact equalities although d_N and d_4 are rational functions of the five-jet; they hold only modulo O(ε²). The density conclusions are unaffected because the leading coefficients are nonzero, but the proofs should state first-order expansions.
[Proof of Proposition 6.13] The reference to 'Proposition 6.12' should be to 'Lemma 6.12', which is the proved bijection between common zeros and ten-jets.
[Lemma 5.9] The phrase 'Sincet∗ V andt ♯ is fixed' contains a typo; it should read 'Since t∗_V and t♯_V are fixed'.
[References] The references [CSW99] and [ISZ02] are listed in the bibliography but are not cited in the body; they should either be cited where relevant or removed from the bibliography.
[Section 1.1] There is a stray period in the sentence introducing V(x) and the homogeneity of V_m; it should read '... where V(x) := V_phys(...), and V_m is homogeneous of degree m in x.'

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the QBNF-to-potential reconstruction is first-principles and self-contained, with only standard external spectral-normal-form equivalences as input.

full rationale

The derivation of Theorem 1.1 (first three QBNF layers determine the Taylor series) is self-contained: the QBNF map is built from Moyal brackets and the finite-dimensional operators A_N, and the injectivity results follow from degree-by-degree triangularity plus Sard-based genericity. The only external inputs are the standard QBNF existence/uniqueness theorem (proved in Appendix B with the cited Weyl calculus [Zwo12]) and the spectral-to-QBNF equivalence of [CdV09], both independent of the present paper's claims. The generic separation step (Proposition 6.7) rests on Lemma 6.5/6.6, whose Fischer-adjoint surjectivity proof is internal and does not presuppose the conclusion. No parameter is fitted to data and then renamed a prediction; the 'admissible value' a12 is recovered by solving a polynomial compatibility system whose bijection with ten-jets is proved (Lemma 6.12), not assumed. The citation of [Wang26] is an acknowledgement of a prior related theorem, not load-bearing, and is not a self-citation by the present author. I therefore find no circular step.

Assumptions & free parameters 0 free parameters · 8 assumptions · 0 invented entities

The central claim is a structural injectivity theorem; it involves no fitted parameters and no new physical entities. The main load-bearing inputs are the irrational-frequency assumption, the Z2 symmetry, the trapped well condition, and an external spectral-to-QBNF equivalence theorem.

assumptions (8)

domain assumption Nonresonance of harmonic frequencies: v1 divided by v2 is irrational.
Used in (1.2) and throughout: it makes the operator L invertible on nonresonant monomials, guarantees uniqueness of the QBNF, and prevents small-denominator collisions such as v1 = 2 v2.
domain assumption Z2 symmetry: V is even in the second normalized coordinate.
Assumed in (1.2); the entire filtration into spaces O_{2N-1} and E_{2N} uses this parity.
domain assumption The leading cubic coefficient a30 has a fixed nonzero sign sigma.
Theorems are stated on the half-space sigma a30 > 0; S3, block inverses, and several generated denominators require a30 nonzero.
domain assumption Unique isolated global well at the origin with compact low sublevel sets, the trapped condition.
Equation (1.3); needed for low-lying eigenvalues to be isolated and for the spectral-to-QBNF bridge.
domain assumption The ordered semiclassical spectrum determines the complete QBNF, as proved by Colin de Verdière and Guillemin-Paul-Uribe.
Invoked in the proof of Theorem 1.4; without this external bridge, spectral equality would not imply QBNF equality.
standard math Formal Moyal calculus: associativity, resonance identities, and QBNF uniqueness.
Proved in Appendix B; the main derivations rely on these identities.
standard math Sard's theorem and Baire category theorem for dense G-delta statements.
Used in Section 6 and in the construction of genericity sets.
standard math Holomorphic-germ topology on real analytic functions as developed by Kriegl and Michor.
Appendix A uses this topology to lift finite-jet generic statements to the analytic class and to prove openness of truncated jet maps.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Generic Spectral Determination of Semiclassical Schr\"odinger Operators with $\mathbb Z_2$-Symmetry." pith.science (2026). https://pith.science/paper/JGCE6EST

@misc{pith2026260811111,
  author       = {Pith},
  title        = {Pith review of: Generic Spectral Determination of Semiclassical Schr\"odinger Operators with $\mathbb Z_2$-Symmetry},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JGCE6EST}},
  note         = {Machine review of arXiv:2608.11111}
}

abstract

Consider the two-dimensional semiclassical Schr\"odinger operator $P_\hbar=-\frac{\hbar^2}{2}\Delta+V$ on $\mathbb R^2$, where $V$ has a nondegenerate well at the origin, its harmonic frequencies are rationally independent, and $V$ is $\mathbb Z_2$-symmetric. We prove that, generically, the first three layers of the quantum Birkhoff normal form (QBNF) determine the full Taylor series of $V$ at the origin, up to the unavoidable spatial inversion $V(x)\mapsto V(-x)$. The generic condition depends only on the jet of $V$ through order ten. Consequently, for real analytic potentials with a unique isolated global well at the origin, the low-lying semiclassical spectrum generically determines $V$ up to spatial inversion.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 25 canonical work pages

[4]

Colin de Verdi\`ere, The semi-classical spectrum and the Birkhoff normal form, arXiv:0902.2470, 2009

Y. Colin de Verdi\`ere, The semi-classical spectrum and the Birkhoff normal form, arXiv:0902.2470, 2009

arXiv 2009
[2]

Cannas da Silva and A

A. Cannas da Silva and A. Weinstein, Geometric Models for Noncommutative Algebras, Berkeley Mathematics Lecture Notes, vol. 10, American Mathematical Society, Providence, RI, 1999

work page 1999
[1]

Borg, Eine Umkehrung der Sturm--Liouvilleschen Eigenwertaufgabe: Bestimmung der Differentialgleichung durch die Eigenwerte, Acta Math

G. Borg, Eine Umkehrung der Sturm--Liouvilleschen Eigenwertaufgabe: Bestimmung der Differentialgleichung durch die Eigenwerte, Acta Math. 78 (1946), 1--96

work page 1946
[3]

Charles and S

L. Charles and S. V \ u Ng o c, Spectral asymptotics via the semiclassical Birkhoff normal form, Duke Math. J. 143 (2008), no. 3, 463--511

work page 2008
[5]

Colin de Verdi\`ere, A semi-classical inverse problem II: reconstruction of the potential, in Geometric Aspects of Analysis and Mechanics, Progress in Mathematics, vol

Y. Colin de Verdi\`ere, A semi-classical inverse problem II: reconstruction of the potential, in Geometric Aspects of Analysis and Mechanics, Progress in Mathematics, vol. 292, Birkh\"auser, Boston, 2011, pp. 97--119

work page 2011
[6]

Colin de Verdi\`ere and V

Y. Colin de Verdi\`ere and V. Guillemin, A semi-classical inverse problem I: Taylor expansions, in Geometric Aspects of Analysis and Mechanics, Progress in Mathematics, vol. 292, Birkh\"auser, Boston, 2011, pp. 81--95

work page 2011
[7]

De Simoi, V

J. De Simoi, V. Kaloshin, and Q. Wei, Dynamical spectral rigidity among $ Z_2$-symmetric strictly convex domains close to a circle (Appendix B coauthored with H. Hezari), Ann.\ of Math. (2) 186 (2017), no. 1, 277--314

work page 2017
[8]

Guillemin and H

V. Guillemin and H. Hezari, A Fulling--Kuchment theorem for the 1D harmonic oscillator, Inverse Problems 28 (2012), no. 4, 045009, 9 pp

work page 2012

Show all 26 references

[9]

Guillemin and A

V. Guillemin and A. Uribe, Some inverse spectral results for semi-classical Schr\"odinger operators, Math. Res. Lett. 14 (2007), no. 4, 623--632

2007
[10]

Guillemin and A

V. Guillemin and A. Uribe, Some inverse spectral results for the two-dimensional Schr\"odinger operator, in Geometry and Analysis, No. 1, Advanced Lectures in Mathematics, vol. 17, International Press, Somerville, MA, 2011, pp. 319--328

2011
[11]

Guillemin, T

V. Guillemin, T. Paul, and A. Uribe, ``Bottom of the well'' semi-classical trace invariants, Math. Res. Lett. 14 (2007), no. 4, 711--719

2007
[12]

Guillemin and Z

V. Guillemin and Z. Wang, Semiclassical spectral invariants for Schr\"odinger operators, J. Differential Geom. 91 (2012), no. 1, 103--128

2012
[13]

Hezari, Inverse spectral problems for Schr\"odinger operators, Comm

H. Hezari, Inverse spectral problems for Schr\"odinger operators, Comm. Math. Phys. 288 (2009), no. 3, 1061--1088

2009
[14]

Hezari and S

H. Hezari and S. Zelditch, One can hear the shape of ellipses of small eccentricity, Ann.\ of Math. (2) 196 (2022), no. 3, 1083--1134

2022
[15]

Iantchenko, J

A. Iantchenko, J. Sj\"ostrand, and M. Zworski, Birkhoff normal forms in semi-classical inverse problems, Math. Res. Lett. 9 (2002), no. 2--3, 337--362

2002
[16]

Kac, Can one hear the shape of a drum?, Amer

M. Kac, Can one hear the shape of a drum?, Amer. Math. Monthly 73 (1966), no. 4, part II, 1--23

1966
[17]

Kriegl and P

A. Kriegl and P. W. Michor, The convenient setting for real analytic mappings, Acta Math. 165 (1990), 105--159

1990
[18]

Levinson, The inverse Sturm--Liouville problem, Mat

N. Levinson, The inverse Sturm--Liouville problem, Mat. Tidsskr. B (1949), 25--30

1949
[19]

V. A. Marchenko, Some questions of the theory of one-dimensional linear differential operators of the second order. I, Trudy Moskov. Mat. Obshch. 1 (1952), 327--420 (Russian)

1952
[20]

Otal, Le spectre marqu\'e des longueurs des surfaces \`a courbure n\'egative, Ann.\ of Math

J.-P. Otal, Le spectre marqu\'e des longueurs des surfaces \`a courbure n\'egative, Ann.\ of Math. (2) 131 (1990), no. 1, 151--162

1990
[21]

V \ u Ng o c, The quantum Birkhoff normal form and spectral asymptotics, Journ\'ees \'Equations aux d\'eriv\'ees partielles (2006), Expos\'e no

S. V \ u Ng o c, The quantum Birkhoff normal form and spectral asymptotics, Journ\'ees \'Equations aux d\'eriv\'ees partielles (2006), Expos\'e no. 10, 12 pp

2006
[22]

Wang, An inverse theorem for partially symmetric two-dimensional semiclassical Schr\"odinger operators, arXiv:2608.04796, 2026

K. Wang, An inverse theorem for partially symmetric two-dimensional semiclassical Schr\"odinger operators, arXiv:2608.04796, 2026

2026 arXiv
[23]

West, A pair of non-isometric potentials with the same semiclassical invariants, J

M. West, A pair of non-isometric potentials with the same semiclassical invariants, J. Math. Phys. 64 (2023), no. 11, 112103

2023
[24]

Zelditch, Inverse spectral problem for analytic domains, II: $ Z_2$-symmetric domains, Ann.\ of Math

S. Zelditch, Inverse spectral problem for analytic domains, II: $ Z_2$-symmetric domains, Ann.\ of Math. (2) 170 (2009), no. 1, 205--269

2009
[25]

Zelditch, Survey on the inverse spectral problem, Notices of the International Congress of Chinese Mathematicians 2 (2014), no

S. Zelditch, Survey on the inverse spectral problem, Notices of the International Congress of Chinese Mathematicians 2 (2014), no. 2, 1--20

2014
[26]

Zworski, Semiclassical Analysis, Graduate Studies in Mathematics, vol

M. Zworski, Semiclassical Analysis, Graduate Studies in Mathematics, vol. 138, American Mathematical Society, Providence, RI, 2012

2012

Pith tools

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

[1] [4]

Colin de Verdi\`ere, The semi-classical spectrum and the Birkhoff normal form, arXiv:0902.2470, 2009

Y. Colin de Verdi\`ere, The semi-classical spectrum and the Birkhoff normal form, arXiv:0902.2470, 2009

arXiv 2009

[2] [2]

Cannas da Silva and A

A. Cannas da Silva and A. Weinstein, Geometric Models for Noncommutative Algebras, Berkeley Mathematics Lecture Notes, vol. 10, American Mathematical Society, Providence, RI, 1999

work page 1999

[3] [1]

Borg, Eine Umkehrung der Sturm--Liouvilleschen Eigenwertaufgabe: Bestimmung der Differentialgleichung durch die Eigenwerte, Acta Math

G. Borg, Eine Umkehrung der Sturm--Liouvilleschen Eigenwertaufgabe: Bestimmung der Differentialgleichung durch die Eigenwerte, Acta Math. 78 (1946), 1--96

work page 1946

[4] [3]

Charles and S

L. Charles and S. V \ u Ng o c, Spectral asymptotics via the semiclassical Birkhoff normal form, Duke Math. J. 143 (2008), no. 3, 463--511

work page 2008

[5] [5]

Colin de Verdi\`ere, A semi-classical inverse problem II: reconstruction of the potential, in Geometric Aspects of Analysis and Mechanics, Progress in Mathematics, vol

Y. Colin de Verdi\`ere, A semi-classical inverse problem II: reconstruction of the potential, in Geometric Aspects of Analysis and Mechanics, Progress in Mathematics, vol. 292, Birkh\"auser, Boston, 2011, pp. 97--119

work page 2011

[6] [6]

Colin de Verdi\`ere and V

Y. Colin de Verdi\`ere and V. Guillemin, A semi-classical inverse problem I: Taylor expansions, in Geometric Aspects of Analysis and Mechanics, Progress in Mathematics, vol. 292, Birkh\"auser, Boston, 2011, pp. 81--95

work page 2011

[7] [7]

De Simoi, V

J. De Simoi, V. Kaloshin, and Q. Wei, Dynamical spectral rigidity among \( Z_2\)-symmetric strictly convex domains close to a circle (Appendix B coauthored with H. Hezari), Ann.\ of Math. (2) 186 (2017), no. 1, 277--314

work page 2017

[8] [8]

Guillemin and H

V. Guillemin and H. Hezari, A Fulling--Kuchment theorem for the 1D harmonic oscillator, Inverse Problems 28 (2012), no. 4, 045009, 9 pp

work page 2012