Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Bounds for D-Algebraic Closure Properties

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · 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 explicit degree bounds for the polynomial differential equations obtained by closure operations on D-algebraic functions, under a complete-intersection condition on the defining equations.

desk verdict Gives the first degree bounds for D-algebraic closure properties, but the advertised bounds hinge on a D-regularity condition that is not shown to be generic, and the written proof contains a repairable Hilbert-series error. read the letter →

arxiv 2505.07304 v1 pith:ZIIWPX5B submitted 2025-05-12 cs.SC

Manuel Kauers , Raphael Pages This is my paper · ORCID
classification cs.SC MSC 12H0513P1068W30
keywords D-algebraicfunctionsdegreeboundsdifferentialeliminationcompleteintersectionHilbertseriesclosurepropertiesorder-degreecurvesalgebra
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

D-algebraic functions, meaning solutions of nonlinear polynomial differential equations, are closed under addition, multiplication, division, and composition. Until now, algorithms for these closure operations could bound the order of the resulting equation but not its degree. This paper proves the first degree bounds, under a genericity condition called D-regularity: when the homogenized derivatives of the defining polynomials form a complete intersection, the elimination ideal must contain a nonzero equation whose degree does not exceed an explicit expression in the input degrees and orders. The bounds are large, and the paper argues this is not mere slack: generic closure can genuinely produce equations that are too big to write out. Sharper resultant-based bounds are proved for the special cases of eliminating algebraic or hyperexponential functions and the variable $x$.

What carries the argument

The carrying object is the tuple of homogenized derivatives $(h(P_j)^{(k)})_{1\le j\le n,\,0\le k\le \rho}$; D-regularity means exactly that this tuple is a complete intersection in the homogenized polynomial ring. For such a tuple the Hilbert series of the ideal it generates is known explicitly, $\prod_i (1-t^{d_i})^{\rho+1}/(1-t)^{r_{\min}+n(\rho+1)}$. The proof compares the coefficient of $t^k$ with the dimension of the space of degree-$k$ polynomials in the variables $s,y_l,\dots,y_l^{(r)}$; once the former is smaller than the latter, some degree-$k$ polynomial must map to zero in the quotient, and setting $s=1$ produces the desired element of the elimination ideal. The resultant arguments in the special cases replace this counting by Sylvester-matrix degree estimates.

What would settle it

Run a complete-intersection test on the homogenized derivative systems of many random D-algebraic inputs of order 2 or 3; because Example 11 already exhibits a D-algebraic system that is not D-regular, the rate of failure is the quantity that decides how widely the bounds apply.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes a conditional size bound for differential elimination. Given differential polynomials $P_1,\dots,P_n$ that define D-algebraic functions and satisfy a D-regularity condition, the elimination ideal $\langle P_1,\dots,P_n\rangle^{(r-r_l)}\cap K[y_l,y'_l,\dots]$ is shown to contain a nonzero element of order $r$ as soon as its degree $k$ exceeds $(r+1)(d^{1+(r_{\min}-r_l)/(r-r_{\min}+1)}-1)$, where $d$ is the product of the degrees of the $P_j$ and $r_{\min}$ is the sum of their orders. From this, the paper derives degree bounds for equations satisfied by $Q(f_1,\dots,f_n)$, covering addition and multiplication, for quotients under an additional joint D-regularity hypothesis, and for compositions, where the bound involves $(r_1+r_2+1)!\,d_2^{r_1}d_1^{r_2}$. It also gives sharper resultant-based bounds for the special cases of eliminating an algebraic function, a hyperexponential function, or the independent variable $x$.

Load-bearing premise

The degree bounds all rest on the assumption that the homogenized derivatives of the defining equations form a complete intersection, the D-regular condition, and the paper does not prove this condition must hold for every D-algebraic system; Example 11 shows it can fail.

Editorial extensions

If this is right

  • For every D-regular tuple, a nonzero differential equation for the selected component of order $r$ exists with degree at most the stated threshold; this is the first degree bound for outputs of D-algebraic closure operations.
  • Sums, products, and more generally $Q(f_1,\dots,f_n)$ inherit the bound whenever each input equation is D-regular, while quotients require a joint D-regularity hypothesis on $(P_1,\dots,P_n,Q_d z-Q_n)$.
  • Compositions satisfy an equation of order $r_1+r_2$ and degree bounded by $(r_1+r_2+1)((r_1+r_2+1)!\,d_2^{r_1}d_1^{r_2}-1)$, and the same equation works for every valid composition.
  • The size of the bounds supports the paper's explanation for why some combinatorial D-algebraic functions have no explicitly written defining equation: generic closure can make the equation prohibitively large.
  • For eliminating an algebraic function, a hyperexponential function, or the variable $x$, resultant-based degree bounds are linear in the input degrees rather than exponential.

Reading between the lines

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

  • If the experimental indication that D-regularity is generic holds, the bound in Theorem 12 is a realistic worst-case estimate; the paper's own comparison with algebraic dependence suggests the counting argument may overshoot by a factor of $r+1$, so a tighter count would immediately improve the bound.
  • The same Hilbert-series device should transfer to differential ideals in several variables or to partial differential equations, where a complete-intersection condition on prolongations would play the same role.
  • The order-degree phenomenon visible in the bound implies a practical algorithmic heuristic: when looking for a defining equation of a D-algebraic combination, trying a few extra derivative orders before raising the target degree can reduce the number of monomials that must be handled.
  • The resultant-based cases indicate that structure-specific elimination can beat the generic bound by orders of magnitude; identifying more such structures is a natural extension.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The manuscript gives degree bounds for the polynomial differential equations obtained from D-algebraic closure properties (addition, multiplication, division, composition). The authors define a D-regularity condition on a tuple (P_1,...,P_n), requiring that the homogenized derivatives h(P_j)^{(k)} form a complete intersection, and prove (Theorem 12) that under this hypothesis the elimination ideal contains a nonzero element of prescribed order r and degree bounded by an expression in the input degrees and orders. The paper then derives corollaries for algebraic operations and composition, and gives separate resultant-based degree bounds for eliminating algebraic functions, hyperexponential functions, and the independent variable x. The authors explicitly acknowledge that D-regularity is nontrivial; Example 11 illustrates a natural D-algebraic system where it fails.

Significance. The paper addresses a natural open question: for D-finite closures both order and degree bounds are known, whereas for D-algebraic closures only order bounds were previously available. If the main bound is correct, the paper provides the first general degree bounds for closure operations on D-algebraic functions. The proof is self-contained, uses standard Hilbert-function and resultant techniques, and contains no fitted parameters. The main limitation is that the advertised bounds are conditional on D-regularity, a hypothesis that is not characterized, is not shown to hold for the systems produced by the known closure algorithms, and can fail on natural examples. The resultant-based bounds of Section 7 are unconditional and are a useful contribution.

major comments (4)
  1. [Section 2, Proposition 7] The stated Hilbert series formula is incorrect as written. The ring R=K[s,x_1,...,x_n] has n+1 variables, so its Hilbert series is (1-t)^{-(n+1)}, not (1-t)^{-n}. For a regular sequence of k homogeneous polynomials, the denominator should be (1-t)^{n+1}. This false identity is later used in the proof of Theorem 12 and should be corrected.
  2. [Section 4, proof of Theorem 12] The displayed Hilbert series for I^{(r-r_l)} has denominator exponent r_min+n(r-r_l+1), but the ambient ring R^h_{r+r_1,...,r+r_n} has one more variable, so the correct denominator exponent is r_min+n(r-r_l+1)+1. The subsequent coefficient bound HF ≤ d^{r-r_l+1} binom(r_min+k,k) can still be justified for the corrected series by the same induction argument, but the manuscript should state this explicitly. As written, the proof of the main theorem contains a false intermediate identity.
  3. [Section 6, proof of Proposition 16] The bound HF_{h(I)}(k) ≤ (r_1+r_2+1)! d_1^{r_2} d_2^{r_1} binom(r_1+r_2+k, r_1+r_2) is stated without derivation. A direct application of the coefficient lemma used in Theorem 12 would give additional factors d_1 d_2, so a sharper coefficient estimate is needed. In addition, the claim that the homogenized family (P_1^{(j)}, P_2^{(j)}, d^i(z-y_1)) is a complete intersection is only asserted by analogy with Proposition 13; the dimension argument should be written out.
  4. [Sections 4 and 5, scope of D-regularity] The main theorems are conditional on D-regularity, but the paper gives no criterion for verifying this condition and does not prove that the systems produced by the closure algorithms of [1] are D-regular. Example 11 shows that a natural D-algebraic system arising from a sum need not be D-regular. Since the title and introduction present the contribution as bounds for closure properties, the authors should state the conditional scope precisely in the abstract and introduction, and discuss whether D-regularity can be checked or enforced.
minor comments (5)
  1. [Section 2, proof of Proposition 8] The proof states that a zero divisor modulo I_i must lie in a minimal prime ideal; this is only immediate after noting that complete intersection ideals are unmixed, and the missing justification should be supplied or replaced by a reference.
  2. [Section 4, Theorem 12] The theorem statement should read 'suppose that (P_1,...,P_n) is D-regular at order r-r_l', and the phrase 'degree k' should be clarified to mean total degree at most k.
  3. [Section 4, Definition 10] The ambient polynomial ring in which the homogenized derivative tuple is required to be a complete intersection should be specified explicitly.
  4. [Figure 1 caption] The caption contains 'r □ rmin'; this should read 'r − r_min'.
  5. [References] Reference [5] contains the typo 'laballed trees'; it should be 'labelled trees'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the degree bounds follow from an explicit Hilbert-function argument under a stated D-regularity hypothesis, not from fitted inputs or self-citations.

full rationale

The paper's central contribution is Theorem 12, which proves a degree bound for elimination ideals under the explicit assumption that the homogenized derivative tuple (h(P_j)^(k)) is a complete intersection (Definition 10). The proof is self-contained: it combines standard Hilbert-series facts (Propositions 5, 7, 8, cited to Cox-Little-O'Shea) with an elementary coefficient bound for products (1+...+t^(d_i-1))^(r-r_l+1), and then concludes by a dimension count in a vector space of homogeneous polynomials. The conclusion is not assumed in the hypothesis and is not a renamed input. The paper expressly acknowledges that D-regularity is a nontrivial condition: Example 11 gives a D-algebraic system that is not D-regular at the required order, and the text says experiments 'seem to indicate that this hypothesis is often satisfied' without claiming a proof. That is an explicitly stated limitation on scope, not a circular derivation. The authors' own prior work appears only for context and comparison (order-degree curves, D-finite closure bounds) and does not carry the argument. No fitted parameter is called a prediction, no uniqueness theorem is imported from self-citations, and no ansatz is smuggled in via citation. Therefore the circularity score is zero.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The main result rests on the D-regularity (complete intersection) assumption, which is the key technical hypothesis, and on standard Hilbert function facts from commutative algebra. One stated Hilbert series formula (Prop 7) is incorrect, but the specific upper bound used can be proved directly. No new entities or fitted parameters are introduced.

assumptions (4)
  • domain assumption The homogenized derivative system (h(P_j)^{(k)}) is a complete intersection (D-regularity).
    This is the technical condition in Definition 10; it is not proven for general D-algebraic ideals and may fail (Example 11). The main theorem and all corollaries depend on it.
  • standard math Hilbert function bound for complete intersections: HF(k) <= (product of degrees) * binom(r_min+k, k).
    Used in Theorem 12. This bound is true by a direct monomial-counting argument, although the paper's Proposition 7 states a stronger exact Hilbert series that is incorrect as written.
  • domain assumption The ideal I is D-algebraic, i.e., the elimination ideal is nonzero after sufficiently many differentiations.
    This is the setup of the paper; it ensures that a D-algebraic equation exists in principle. It is part of the definition of D-algebraic ideals in Section 3.
  • domain assumption Composition is defined abstractly via Definition 15.
    Proposition 16's result depends on this algebraic definition of composition, which generalizes the usual function composition but is not the only possible formalization.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bounds for D-Algebraic Closure Properties." pith.science (2026). https://pith.science/paper/ZIIWPX5B

@misc{pith2026250507304,
  author       = {Pith},
  title        = {Pith review of: Bounds for D-Algebraic Closure Properties},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZIIWPX5B}},
  note         = {Machine review of arXiv:2505.07304}
}
read the original abstract

We provide bounds on the size of polynomial differential equations obtained by executing closure properties for D-algebraic functions. While it is easy to obtain bounds on the order of these equations, it requires some more work to derive bounds on their degree. Here we give bounds that apply under some technical condition about the defining differential equations.

Figures

Figures reproduced from arXiv: 2505.07304 by the authors.

Figure 1
Figure 1. order-number of monomials curves It was not obtained by actually solving these linear systems, for they are too big to handle, so these curves may overshoot. 5 Corollaries on degree bounds for algebraic operations Proposition 13. Let 𝑓1, . . . , 𝑓𝑛 be D-algebraic functions, as well as 𝑃1, . . . , 𝑃𝑛 ∈ 𝑅𝑟1,...,𝑟𝑛 such that 𝑃𝑖 ∈ 𝐾[𝑦𝑖 , 𝑦′ 𝑖 , . . . , 𝑦 (𝑟𝑖 ) 𝑖 ] for all 𝑖 ∈ {1, . . . , 𝑛} and 𝑄 ∈ 𝑅𝑟1,...,𝑟𝑛 . We note … view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Support bound for differential elimination in polynomial dynamical systems

    cs.SC 2025-06 conditional novelty 6.0 of 10

    A support bound for the minimal differential equation of a polynomial observation of a polynomial dynamical system, generalized to arbitrary polynomial outputs and symbolic parameters.

Reference graph

Works this paper leans on

27 extracted references · 23 canonical work pages · cited by 1 Pith paper

  1. [1]

    Rida Ait El Manssour, Anna-Laura Sattelberger, and Bertrand Teguia Tabuguia

    work page

  2. [2]

    Olivier Bernardi and Mireille Bousquet-Mélou. 2017. Counting Coloured Planar Maps: Differential Equations.Communications in Mathematical Physics354 (2017), 31–84

    work page 2017

  3. [3]

    Oliver Bernardi, Mireille Bousquet-Mélou, and Kilian Raschel. 2021. Counting quadrant walks via Tutte’s invariant method.Combinatorial Theory1 (2021), #3

    work page 2021

  4. [4]

    Alin Bostan, Frédéric Chyzak, Bruno Salvy, Grégoire Lecerf, and Éric Schost

    work page

  5. [5]

    Alin Bostan and Antonio Jiménez-Pastor. 2020. On the exponential generating function of laballed trees.Comptes Rendus Mathématique358, 9–10 (2020), 1005– 1009

    work page 2020

  6. [6]

    Francois Boulier. 1996. An optimization of Seidenberg’s elimination algorithm in differential algebra.Mathematics and Computers in Simulation42 (1996), 439–448

    work page 1996

  7. [7]

    Mireille Bousquet-Mélou and Andrew Elvey Prince. 2022. The generating function of planar Eulerian orientations.Journal of Combinatorial Theory A172 (2022), 105183

    work page 2022

  8. [8]

    Carra-Ferro

    G. Carra-Ferro. 1997. A resultant theory for the systems of two ordinary algebraic differential equations.Applicable Algebra in Engineering, Communications and Computing8, 6 (1997), 539–560

    work page 1997

Show all 27 references
  1. [9]

    Carra-Ferro

    G. Carra-Ferro. 2007. A survey on differential Gröbner bases. InGröbner bases in Symbolic Analysis. De Gruyter, 77–108

    2007

  2. [10]

    Shaoshi Chen, Hanqian Fang, Sergey Kitaev, and Candice X.T. Zhang. 2024. Patterns in Multi-dimensional Permutations. Technical Report 2411.02897. ArXiv

    2024 arXiv

  3. [11]

    Shaoshi Chen, Maximilian Jaroschek, Manuel Kauers, and Michael F. Singer. 2013. Desingularization Explains Order-Degree Curves for Ore Operators. InProc. ISSAC’13. ACM, 157–164

    2013

  4. [12]

    Shaoshi Chen and Manuel Kauers. 2012. Order-Degree Curves for Hypergeomet- ric Creative Telescoping. InProceedings of ISSAC’12. ACM, 122–129

    2012

  5. [13]

    2007.Ideals, Varieties, and Algorithms

    David Cox, John Little, and Donal O’Shea. 2007.Ideals, Varieties, and Algorithms. An Introduction to Computational Algebraic Geometry and Commutative Algebra. Springer. https://link.springer.com/book/10.1007/978-0-387-35651-8

    2007 doi

  6. [14]

    Denef and L

    J. Denef and L. Lipshitz. 1984. Power Series Solutions of Algebraic Differential Equations.Math. Ann.267 (1984), 213–238

    1984

  7. [15]

    Denef and L

    J. Denef and L. Lipshitz. 1989. Decision Problems for Differential Equations.The Journal of Symbolic Logic54, 3 (1989), 941–950

    1989

  8. [16]

    Manuel Kauers. 2014. Bounds for D-finite Closure Properties. InProceedings of ISSAC’14. ACM, 288–295

    2014

  9. [17]

    2023.D-Finite Functions

    Manuel Kauers. 2023.D-Finite Functions. Springer

    2023

  10. [18]

    E. R. Kolchin. 1973.Differential Algebra and Algebraic Groups. Academic Press

    1973

  11. [19]

    2025.Projecting dynamical systems via a support bound

    Yulia Mukhina and Gleb Pogudin. 2025.Projecting dynamical systems via a support bound. Technical Report 2501.13680. ArXiv

    2025 arXiv

  12. [20]

    1991.Standard bases of differential ideals

    Francois Ollivier. 1991.Standard bases of differential ideals. Springer

    1991

  13. [21]

    Josef F. Ritt. 1950.Differential Algebra. American Mathematical Society, Collo- quium Publications

    1950

  14. [22]

    Sonia Rueda. 2016. Differential elimination by differential specialization of Sylvester style matrices.Advances in Applied Mathematics72 (2016), 4–37

    2016

  15. [23]

    Seidenberg

    A. Seidenberg. 1956. An elimination theory for differential algebra.Univ. Califor- nia Publ. MathIII, 22 (1956), 31–38

    1956

  16. [24]

    Shafarevich and Miles Reid

    Igor R. Shafarevich and Miles Reid. 1994.Basic algebraic geometry 1 (2nd, revised and expanded ed.). Springer-Verlag, Berlin, Heidelberg

    1994

  17. [25]

    Joris van der Hoeven. 2019. Computing with D-algebraic power series.AAECC 30 (2019), 17–49

    2019

  18. [2007]

    InProceedings of ISSAC’07

    Differential Equations for Algebraic Functions. InProceedings of ISSAC’07. ACM, 25–32

  19. [2025]

    https://doi.org/10.1016/j.jsc.2024.102377

    D-algebraic functions.Journal of Symbolic Computation128 (2025), 102377. https://doi.org/10.1016/j.jsc.2024.102377

    2025

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.