Pith. sign in

REVIEW 5 minor 33 references

This paper proves a refined counting bound for Pfaffian systems: the number of regular real solutions is controlled by k, the number of variables the Pfaffian chain actually depends on, rather than the ambient dimension n.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

A self-contained proof of Khovanskii's Bezout-type bound for Pfaffian systems, refined to depend on the number k of variables in the chain.

T0 review reviewed 2026-08-03 challenge →

load-bearing objection A genuinely self-contained proof of Khovanskii's Pfaffian Bezout bound with a real but mild improvement when the chain depends on few variables; the referee should engage.

arxiv 2607.29267 v1 pith:IOQII4QR submitted 2026-07-31 math.AG math.CAmath.LO

Khovanskii's Bezout-type Theorem for Pfaffian Functions: A Self-Contained Proof, and Applications

Martin Lotz , Abhiram Natarajan This is my paper
classification math.AG math.CAmath.LO MSC 14P1014P15
keywords Pfaffian functionsPfaffian chainscounting boundcomparison lemmaconnected componentsBetti numbersreal algebraic geometryo-minimal structures
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

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 the classical counting bound for Pfaffian systems and refines it: the number of regular real solutions depends on k, the number of variables on which the Pfaffian chain actually depends, rather than the ambient dimension n. The refined bound replaces the factor min{n,s} in the classical statement with min{k,s}. A sympathetic reader should care because short or low-dimensional chains are common in practice, and the sharper count propagates directly into bounds on connected components of Pfaffian sets. The proof is a self-contained induction on the chain order, built around a comparison step that lowers the order by one while paying only for the intrinsic dimension of the chain.

Core claim

The paper's central claim is that the number of regular real solutions of f_1=...=f_n=0 is at most β_1...β_n · 2^{s(s-1)/2} · (Σβ_i - n + min{k,s}α + 1)^s, where (α,β_i,s) is the format of the Pfaffian functions and the chain depends only on k≤n variables. The refinement is the replacement of min{n,s}α by min{k,s}α. The decisive observation is that, after replacing the last chain function q_s with a new variable x_{n+1}, the determinant used in the comparison becomes Pfaffian of order s-1 whose degree is controlled by k; the induction on s then closes.

What carries the argument

Central is the (n+1)-by-(n+1) matrix M=[∇g_1,...,∇g_n,η], where η is obtained from the gradient of g=q_s(x)-x_{n+1} by substituting x_{n+1} for q_s inside the chain polynomials. The determinant-degree estimate (Lemma 3.3) bounds the degree of det(M) by Σ(β_i-1)+min{k,s}α. The comparison lemma then turns the original solution count into a count of points where det(M) equals a small regular value, and because det(M) belongs to the shorter chain (q_1,...,q_{s-1}), the induction on chain order can proceed.

Load-bearing premise

The proof stands on the determinant-degree estimate: after replacing q_s by x_{n+1}, the comparison determinant has degree at most Σ(β_i-1)+min{k,s}α; if the cofactor bound min{k-1,s-1}+1=min{k,s} fails, the induction cannot close and the final exponent reverts to n.

What would settle it

Compute det(M) for a concrete Pfaffian system with chain depending on k variables and check whether its degree exceeds Σ(β_i-1)+min{k,s}α. For example, with s=2, k=2 and β_i=1, the claim says the cofactor of h_{2,j} has degree at most 2α; expanding it symbolically and finding any monomial of degree greater than 2α would disprove the lemma.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • When k=n, the refined bound reduces to the classical statement; when k<n it is strictly smaller, so all existing estimates using n improve automatically.
  • The connected-component bound for Pfaffian sets inherits the same k-dependence, making it stronger for chains of order s=o(n) or chains using few variables.
  • The proof is self-contained and does not rely on the general theory of integral manifolds, so the mechanism behind the bound is directly inspectable.
  • Non-proper systems are handled by adding a sphere equation, which shows the bound is independent of truncation and remains valid on all of R^n.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same replacement of min{n,s} by min{k,s} most likely carries over to semi-Pfaffian and sub-Pfaffian Betti-number estimates; the paper only spells out the connected-component case, but the mechanism is not limited to that invariant.
  • A testable consequence is that, for chains depending on few variables, solution counts should be dramatically smaller; constructing explicit families of Pfaffian systems with k fixed and n large and measuring the maximum number of solutions would probe whether the min{k,s} factor is tight.
  • The paper notes in Remark 2.3 that its proof is written on R^n; extending the argument to the usual non-compact domains requires a non-compact version of the comparison step, which the authors leave open.
  • The block-determinant expansion in Lemma 3.3 is abbreviated; a fully written expansion of the cofactors involving h_{s,j} is the place to check the degree estimate before porting the method to other settings.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 5 minor

Summary. The paper presents a self-contained proof of Khovanskii's Bézout-type bound for systems of Pfaffian equations. The main theorem refines the classical statement by replacing the ambient dimension n with k, the number of variables on which the Pfaffian chain actually depends, in the exponent base. The proof proceeds by induction on the chain order, isolating a Khovanskii–Rolle lemma, a determinant degree bound (Lemma 3.3), and a properness-removal argument. As an application, the authors derive a bound on the number of connected components of a Pfaffian set.

Significance. If correct, the paper provides a useful and genuinely self-contained proof of a central result in Pfaffian geometry, and the k-refinement is a mild but real strengthening: it can improve bounds in settings where the chain depends on few variables, e.g., when s=o(n) or k=o(n). The proof is detailed, with explicit constants and no fitted parameters. The main lemmas, including the Khovanskii–Rolle step and the determinant-degree estimate, are proved in full, and the application to connected components is a natural illustration. The reader's and skeptic's checks confirm that the induction closes and the refined degree bound is correct.

minor comments (5)
  1. [Theorem 1.2 (Section 1 and Section 5)] The zero set in the statement is written as {g_1(x)=...=g_k(x)=0}; this should be {f_1(x)=...=f_m(x)=0}. In the proof, the notation F=(f_1,...,f_m) is introduced without definition. Please correct these typos.
  2. [Lemma 3.3] The definition h_{s,j}:=P_{s,j}(x_1,...,x_k,q_1,...,q_{s-1},x_{n+1}) implicitly restricts the polynomial P_{s,j}, which originally has n X-variables, to the first k variables. Since the chain depends only on x_1,...,x_k, one may replace P_{s,j} by its restriction to those variables without changing the chain equations; this reduction should be stated explicitly to avoid ambiguity.
  3. [Lemma 3.3] The cofactor computation for the terms involving h_{s,j} is very terse. The sentence 'expanding along it and then along the resulting last row shows...' skips the key counting step that yields the min{k-1,s-1}α contribution. Since this is the load-bearing step for the k-refinement, the derivation should be expanded for readability and verification.
  4. [Claim 4.2] The variable naming is inconsistent: for a system on R^{n+1}, Step 1 of the proof introduces x_{n+2} in the general notation, but Claim 4.2 calls the new variable x_{n+1}. This can confuse the reader; please clarify the variable indexing in the lifted map.
  5. [Remark 2.3] The remark notes that the proof is restricted to U=R^n and that the bound remains valid on other domains only after additional work. This is fine, but it would be helpful to state explicitly that Theorem 1.1 is proved only for functions defined on all of R^n, matching the theorem statement.

Circularity Check

0 steps flagged

No significant circularity: the proof is a self-contained induction and the k-refinement emerges from the degree count.

full rationale

Theorem 1.1 is proved by induction on the Pfaffian chain order s, with the target bound used only as the induction hypothesis for smaller s; it is not assumed for the case being proved. The base case s=0 is exactly the classical real Bezout inequality (Theorem 2.9), and the Khovanskii-Rolle machinery (Lemma 3.1, Corollary 3.2) is proved in the paper for arbitrary smooth maps, not imported as a black box. The only Pfaffian-specific ingredient is Lemma 3.3, which bounds the degree of det(M) by an explicit block-determinant/Laplace-expansion computation using the displayed derivative formula (3.3) and the fact that P_ell,j = 0 for j > k; the min{k,s} alpha term comes from counting nonzero rows in U, not from a prefitted parameter. The authors' own cited works appear as illustrations or applications, not as load-bearing inputs: closure properties are also referenced to Zel03b, and the proof does not invoke the classical Khovanskii bound. Theorem 1.2 applies the already-proved Theorem 1.1 to a Milnor-type critical-point reduction, which is a standard consequence, not a circular one. The restriction to U = R^n (Remark 2.3) and the compressed cofactor computation in Lemma 3.3 are expositional gaps, but they do not make any step equivalent by construction to its input.

Axiom & Free-Parameter Ledger

0 free parameters · 7 axioms · 0 invented entities

The proof uses only standard theorems (Real Bezout, Sard, inverse function theorem, basic 1-manifold facts) plus the defining assumptions of Pfaffian functions. There are no fitted constants and no invented entities. The k-dependence comes from a harmless convention: chain polynomials for variables the chain does not depend on can be chosen to be zero. The only domain assumption that could limit scope is the restriction to functions defined on all of R^n, which the paper flags in Remark 2.3.

axioms (7)
  • standard math Real Bezout inequality: number of nondegenerate solutions of n polynomial equations in n variables is at most the product of the degrees.
    Used in the base case s=0 and stated as Theorem 2.9 with a proof via [BCR13, Lemma 11.5.1].
  • standard math Sard's theorem: critical values have measure zero, so regular values exist.
    Used to choose generic y, delta, epsilon, and c throughout the proof (Steps 2-3 and Theorem 1.2).
  • standard math Inverse function theorem.
    Used in Step 2 to perturb solutions and in Claim 4.1 to relate regular solutions of F and of (G,g).
  • standard math Compact connected 1-manifold is diffeomorphic to S^1.
    Used in Corollary 3.2 to count sign changes of g on components of C.
  • standard math Nonvanishing vector field on a 1-manifold admits a parameterization by its flow.
    Used in the proof of Lemma 3.1 to parameterize the arc C_0; cited to [Tu11, Chapter 14].
  • domain assumption Pfaffian functions and the Pfaffian chain are defined on all of R^n.
    The proof of Theorem 1.1 uses global properness constructions; Remark 2.3 explicitly notes that extension to other domains requires additional work on non-compact curves.
  • domain assumption Chain polynomials for variables outside {x_1,...,x_k} may be taken to be zero.
    Needed in Lemma 3.3 to assert P_{\ell,j}=0 for j>k and to handle the new variables x_{n+1} and x_0. This follows from the chain functions being independent of those variables, by choosing a suitable chain representation.

reviewed 2026-08-03 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Khovanskii's Bezout-type Theorem for Pfaffian Functions: A Self-Contained Proof, and Applications." pith.science (2026). https://pith.science/paper/IOQII4QR

@misc{pith2026260729267,
  author       = {Pith},
  title        = {Pith review of: Khovanskii's Bezout-type Theorem for Pfaffian Functions: A Self-Contained Proof, and Applications},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IOQII4QR}},
  note         = {Machine review of arXiv:2607.29267}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We present a direct and self-contained proof of Khovanskii's Bezout-type bound for the number of nondegenerate solutions of a system of Pfaffian equations. We isolate the ingredients of Khovanskii's original argument and assemble them into a proof that avoids the general theory of integral manifolds developed in his monograph. Our formulation mildly refines the classical statement: rather than depending on the ambient dimension, our bound depends on the maximum number of variables on which any function in the Pfaffian chain depends. As a consequence, we obtain a refined bound on the number of connected components of a Pfaffian set.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

33 extracted references · 1 canonical work pages

  1. [1]

    and Vorobjov, N

    Gabrielov, A. and Vorobjov, N. and Zell, T. , title =. Journal of the London Mathematical Society , volume =. doi:https://doi.org/10.1112/S0024610703004939 , year =

    work page doi:10.1112/s0024610703004939

  2. [2]

    2025 , eprint=

    Distinct Distances on Pfaffian Curves , author=. 2025 , eprint=

    2025

  3. [3]

    2026 , eprint=

    Tubular Neighbourhoods of Pfaffian Sets and Applications to Neural Networks , author=. 2026 , eprint=

    2026

  4. [4]

    Parameterwise Sharpness of

    Terence Bickerton and Joseph Harrison and Olivia Hornakova and Dominic Le-Mar and Abhiram Natarajan and Nadia Potter , year=. Parameterwise Sharpness of. 2606.24373 , note=

    Pith/arXiv arXiv

  5. [5]

    Khovanskii's Theorem

    Marker, David. Khovanskii's Theorem. Algebraic Model Theory. 1997. doi:10.1007/978-94-015-8923-9_8

    doi:10.1007/978-94-015-8923-9_8 1997

  6. [6]

    IEEE transactions on neural networks and learning systems , volume=

    On the complexity of neural network classifiers: A comparison between shallow and deep architectures , author=. IEEE transactions on neural networks and learning systems , volume=. 2014 , publisher=

    2014

  7. [7]

    Quarterly journal of mathematics , volume=

    The density of algebraic points on certain Pfaffian surfaces , author=. Quarterly journal of mathematics , volume=. 2012 , publisher=

    2012

  8. [8]

    Partitioning theorems for sets of semi-Pfaffian sets, with applications , volume=

    Lotz, Martin and Natarajan, Abhiram and Vorobjov, Nicolai , year=. Partitioning theorems for sets of semi-Pfaffian sets, with applications , volume=. doi:10.1017/fms.2025.10150 , journal=

    arXiv 2025

  9. [9]

    2003 , school=

    Quantitative study of semi-Pfaffian sets , author=. 2003 , school=

    2003

  10. [10]

    Foundations of Computational Mathematics , volume=

    Khovanskii--Rolle continuation for real solutions , author=. Foundations of Computational Mathematics , volume=. 2011 , publisher=

    2011

  11. [11]

    Quantitative study of semi-Pfaffian sets

    Thierry Paul Zell. Quantitative study of semi-Pfaffian sets. 2003

    2003

  12. [12]

    Milnor, John , journal=. On the

  13. [13]

    Sur l’homologie des vari

    Thom, Ren. Sur l’homologie des vari. Differential and combinatorial topology , pages=

  14. [14]

    Izvestiya Rossiiskoi Akademii Nauk

    On the topology of real algebraic surfaces , author=. Izvestiya Rossiiskoi Akademii Nauk. Seriya Matematicheskaya , volume=. 1949 , publisher=

    1949

  15. [15]

    1998 , publisher=

    Tame topology and o-minimal structures , author=. 1998 , publisher=

    1998

  16. [16]

    Discrete & Computational Geometry , volume=

    Counting real connected components of trinomial curve intersections and m -nomial hypersurfaces , author=. Discrete & Computational Geometry , volume=. 2003 , publisher=

    2003

  17. [17]

    New fewnomial upper bounds from

    Bihan, Fr. New fewnomial upper bounds from. Moscow Mathematical Journal , volume=

  18. [18]

    2011 , publisher=

    Real solutions to equations from geometry , author=. 2011 , publisher=

    2011

  19. [19]

    2002 , publisher=

    A Primer of Real Analytic Functions , author=. 2002 , publisher=

    2002

  20. [20]

    Normal forms, bifurcations and finiteness problems in differential equations , volume=

    Complexity of computations with Pfaffian and Noetherian functions , author=. Normal forms, bifurcations and finiteness problems in differential equations , volume=. 2004 , publisher=

    2004

  21. [21]

    Miller, Chris and Speissegger, Patrick , TITLE =. J. Symbolic Logic , FJOURNAL =. 2002 , NUMBER =

    2002

  22. [22]

    Lion, Jean-Marie and Rolin, Jean-Philippe , TITLE =. Ann. Fac. Sci. Toulouse Math. (6) , FJOURNAL =. 1998 , NUMBER =

    1998

  23. [23]

    1997 , publisher=

    Topology from the Differentiable Viewpoint , author=. 1997 , publisher=

    1997

  24. [24]

    2011 , publisher=

    An Introduction to Manifolds , author=. 2011 , publisher=

    2011

  25. [25]

    2013 , publisher=

    Real algebraic geometry , author=. 2013 , publisher=

    2013

  26. [26]

    Selecta Mathematica , volume=

    A theorem of the complement and some new o-minimal structures , author=. Selecta Mathematica , volume=. 1999 , publisher=

    1999

  27. [27]

    Journal of the American Mathematical Society , volume=

    Model completeness results for expansions of the ordered field of real numbers by restricted Pfaffian functions and the exponential function , author=. Journal of the American Mathematical Society , volume=

  28. [28]

    Speissegger, Patrick , TITLE =. J. Reine Angew. Math. , FJOURNAL =. 1999 , PAGES =

    1999

  29. [29]

    New fewnomial upper bounds from Gale dual polynomial systems , journal =

    Bihan, Fr. New fewnomial upper bounds from Gale dual polynomial systems , journal =

  30. [30]

    Betti numbers of semi-

    Zell, Thierry , journal=. Betti numbers of semi-

  31. [31]

    1991 , publisher=

    Fewnomials , author=. 1991 , publisher=

    1991

  32. [32]

    , title =

    Khovanskii, Askold G. , title =. Doklady Akademii Nauk , volume =

  33. [33]

    , title =

    Khovanskii, Askold G. , title =. Proceedings of the. 1984 , isbn =

    1984

This paper was first reviewed by deepseek-v4-flash on August 3, 2026.