On Nests and Large Components of Random Real Algebraic Curves

Ali Ula\c{s} \"Ozg\"ur Ki\c{s}isel; Turgay Bayraktar

arxiv: 2604.18350 · v1 · submitted 2026-04-20 · 🧮 math.AG · math.PR

On Nests and Large Components of Random Real Algebraic Curves

Ali Ula\c{s} \"Ozg\"ur Ki\c{s}isel , Turgay Bayraktar This is my paper

Pith reviewed 2026-05-10 03:39 UTC · model grok-4.3

classification 🧮 math.AG math.PR

keywords Kostlan polynomialsrandom real algebraic curvesbarrier methodconnected componentsnestsplane curvesexpected topologycomplement components

0 comments

The pith

In high-degree Kostlan random real algebraic curves, the expected number of long connected components and deep nests both tend to infinity.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The authors adapt the barrier method to study topological features of random plane curves defined by Kostlan polynomials. They establish that the expected count of curve components whose length exceeds order sqrt(log log d over d) diverges as the degree d grows. They prove an analogous divergence for the expected number of nests whose depth exceeds order log log d. A separate argument adapts an L-infinity norm bound to obtain a positive lower bound on the probability that finitely many prescribed points lie in distinct connected components of the complement. These results describe how the typical topology of such random curves becomes richer with increasing degree.

Core claim

We develop a variant of the barrier method in order to address questions about topology of Kostlan random real algebraic plane curves. In particular we prove that the expected number of connected components of the curve of length at least O(sqrt(d^{-1} log log d)) grows to infinity with d, and likewise, the expected number of nests of the curve of depth at least O(log log d) grows to infinity with d. We also adapt an L^infty-norm bound result of Shifman and Zelditch to subspaces and employ it to obtain a lower bound for the probability that a finite number of points remain all in different components of the complement of a large degree random curve.

What carries the argument

Variant of the barrier method applied to level sets of Kostlan polynomials, used to control the occurrence of long arcs and deep nestings.

If this is right

The typical random curve of degree d is expected to contain more and more macroscopic connected components as d increases.
Nesting structures in the complement of the curve become deeper in expectation.
The probability that any fixed finite set of points is separated by the random curve remains bounded away from zero.
Topological complexity measured by component length and nesting depth diverges in the high-degree limit.

Where Pith is reading between the lines

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

The divergence results suggest that the model may exhibit percolation-like behavior for topological features at certain scales.
Similar barrier techniques could be tested on random polynomials with other coefficient distributions or in higher dimensions.
The separation probability bound may connect to questions about the typical distance between components in the complement.

Load-bearing premise

The authors' variant of the barrier method applies directly to the level sets of Kostlan polynomials without hidden restrictions that invalidate the stated length and depth thresholds.

What would settle it

An explicit upper bound or numerical computation showing that the expected number of components of length at least sqrt(log log d / d) stays bounded for arbitrarily large d.

read the original abstract

We develop a variant of the barrier method in order to address questions about topology of Kostlan random real algebraic plane curves. In particular we prove that the expected number of connected components of the curve of length at least $\displaystyle{O\left(\sqrt{d^{-1}\log \log d}\right)}$ grows to infinity with $d$, and likewise, the expected number of nests of the curve of depth at least $\displaystyle{O\left(\log\log d\right)}$ grows to infinity with $d$. In another direction, we adapt an $L^{\infty}$-norm bound result of Shifmann and Zelditch to subspaces and employ it to obtain a lower bound for the probability that a finite number of points remain all in different components of the complement of a large degree random curve.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper proves diverging expectations for long components and deep nests in Kostlan plane curves at explicit log-log scales using a barrier variant, plus a subspace adaptation of the Shifman-Zelditch bound.

read the letter

The core new results are that the expected number of connected components of length at least O(sqrt(d^{-1} log log d)) tends to infinity with d, and the expected number of nests of depth at least O(log log d) does the same. They obtain these by constructing a variant of the barrier method for the level sets of Kostlan polynomials and by extending the Shifman-Zelditch L^infty bound to work on finite-dimensional subspaces, which gives a lower bound on the probability that a fixed number of points lie in distinct complementary components of a high-degree random curve. Both statements are new at this level of explicitness; earlier work on random real curves had mostly produced existence or upper-bound statements rather than these diverging lower bounds on counts. The technical steps look like a reasonable adaptation of existing tools to the Kostlan setting. The main soft spot is whether the barrier variant really produces events with positive probability that are independent enough at the precise log-log thresholds. Kostlan fields are global, with covariance fixed by the reproducing kernel, so any barrier construction needs to confirm that the microscopic length and depth scales survive the global constraints and variance normalization without losing the divergence. The abstract asserts that the proofs go through, but the stress-test concern about covariance controls is worth checking in the details; if the thresholds hold only up to a constant factor it would still be useful but weaker than stated. This is a specialized paper aimed at people who work on random algebraic geometry or the topology of Gaussian fields on projective space. It is a clear incremental step with concrete claims, so it deserves a serious referee who can verify the barrier construction and the subspace bound. I would send it to review rather than desk-reject.

Referee Report

2 major / 2 minor

Summary. The manuscript develops a variant of the barrier method to study the topology of Kostlan random real algebraic plane curves. It proves that the expected number of connected components of length at least O(sqrt(d^{-1} log log d)) grows to infinity with d, and that the expected number of nests of depth at least O(log log d) likewise diverges. It further adapts an L^∞-norm bound of Shifman and Zelditch to subspaces, yielding a lower bound on the probability that a finite collection of points lies in distinct components of the complement of a large-degree random curve.

Significance. If the claims hold, the work supplies quantitative lower bounds on the typical appearance of macroscopic topological features (long components and deep nests) in random real plane curves of growing degree. The barrier-method variant and the subspace-adapted norm bound are potentially reusable tools for random algebraic geometry and Gaussian random fields on projective space.

major comments (2)

[barrier-method variant] Barrier-method section: the custom barrier construction is asserted to produce positive-probability events for components of length ≳ sqrt(d^{-1} log log d) and nests of depth ≳ log log d, yet the text does not contain an explicit verification that the log-log corrections survive the global reproducing-kernel covariance and variance normalization of the Kostlan ensemble; without such controls the divergence of the expectations does not necessarily follow.
[L^∞-norm adaptation] Adaptation of Shifman-Zelditch L^∞ bound: the extension to subspaces is invoked to obtain the separation-probability lower bound, but the argument lacks a precise statement of the subspace choice and a uniformity check that the bound remains valid at the finite sets of points under consideration.

minor comments (2)

[Introduction] The definitions of 'nest' and 'depth' should be stated explicitly at the first appearance rather than deferred to later sections.
[Abstract and main theorems] Notation for the Kostlan ensemble and the scaling of the length and depth thresholds should be made uniform between the abstract and the main statements.

Simulated Author's Rebuttal

2 responses · 0 unresolved

Thank you for the detailed referee report. We appreciate the positive assessment of the significance of our results. Below we respond to the major comments, indicating the revisions we will make to address them.

read point-by-point responses

Referee: Barrier-method section: the custom barrier construction is asserted to produce positive-probability events for components of length ≳ sqrt(d^{-1} log log d) and nests of depth ≳ log log d, yet the text does not contain an explicit verification that the log-log corrections survive the global reproducing-kernel covariance and variance normalization of the Kostlan ensemble; without such controls the divergence of the expectations does not necessarily follow.

Authors: We thank the referee for pointing this out. Upon review, we agree that an explicit verification is necessary to rigorously confirm that the log-log corrections persist under the Kostlan ensemble's global covariance and normalization. In the revised manuscript, we will add a new paragraph or subsection in the barrier-method section that computes the relevant covariance bounds using the explicit form of the reproducing kernel for the Kostlan measure on the projective plane. This will show that the variance normalization affects the probabilities only by constant factors, preserving the divergence of the expectations as d → ∞. revision: yes
Referee: Adaptation of Shifman-Zelditch L^∞ bound: the extension to subspaces is invoked to obtain the separation-probability lower bound, but the argument lacks a precise statement of the subspace choice and a uniformity check that the bound remains valid at the finite sets of points under consideration.

Authors: We acknowledge the need for greater precision here. The subspace in question is the finite-dimensional subspace spanned by the monomials corresponding to the points in the finite collection, but we will make this explicit. Additionally, we will include a uniformity check by applying the Shifman-Zelditch bound with constants that are uniform over the choice of points in a compact set away from the curve's singularities or using the fact that for large d the points are separated. This will be added to the relevant section on the L^∞-norm adaptation. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation uses external results and independent variant

full rationale

The paper develops a variant of the barrier method and adapts an external L^∞-norm bound from Shifman and Zelditch to subspaces. The main results on expected numbers of large components and deep nests growing to infinity are obtained via probabilistic estimates on Kostlan ensembles. No self-definitional steps, fitted inputs renamed as predictions, or load-bearing self-citations appear; the derivation chain relies on independent method development and external benchmarks rather than reducing to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper rests on the standard Kostlan probability measure and the applicability of a modified barrier method; no free parameters or new entities are introduced.

axioms (2)

domain assumption Kostlan distribution defines the natural rotationally invariant Gaussian measure on the space of degree-d real polynomials.
This is the standard model for random real algebraic curves invoked throughout the abstract.
domain assumption A variant of the barrier method can be constructed to control topological features of the zero sets of these random polynomials.
The central technical step of the paper.

pith-pipeline@v0.9.0 · 5439 in / 1409 out tokens · 52050 ms · 2026-05-10T03:39:36.257514+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

[1]

The Bergman kernel and a theorem of Tian

Catlin, D. “The Bergman kernel and a theorem of Tian” Analysis and geometry in several complex variables, 1–23, Trends Math., Birkhauser (1999)

work page 1999
[2]

On the∂equation in weightedL 2 notms inC 1

Christ, M. “On the∂equation in weightedL 2 notms inC 1” J. Geometric Anal., 1, 193–230 (1991)

work page 1991
[3]

Low degree approximation of random polynomials

Diatta, D. N., Lerario, A. “Low degree approximation of random polynomials” Foundations of Computational Math., Vol. 22, 77–97 (2022)

work page 2022
[4]

How many zeros of a random polynomials are real?

Edelman, A. , Kostlan, E. “How many zeros of a random polynomials are real?” Bull. Amer. Math. Soc. 32, no. 1, 1–37 (1995)

work page 1995
[5]

On the number of connected components of random algebraic hypersurfaces

Fyodorov, Y. V., Lerario, A., Lundberg, E. “On the number of connected components of random algebraic hypersurfaces” J. Geometry and Physics, Vol. 95, 1–20 (2015)

work page 2015
[6]

Betti numbers of random real hypersurfaces and determinants of random symmetric matrices

Gayet, D. , Welschinger, J-Y. “Betti numbers of random real hypersurfaces and determinants of random symmetric matrices”, J. European Matth. Soc., Vol. 18, 732–772 (2016)

work page 2016
[7]

Lower estimates for the expected Betti numbers of random real hypersurfaces

Gayet, D. , Welschinger, J-Y. “Lower estimates for the expected Betti numbers of random real hypersurfaces”, J. London Math. Soc. Vol. 90, 105–120 (2014)

work page 2014
[8]

Amoeba measures of random plane curves

Kişisel, A. U. Ö. , Welschinger, J-Y. “Amoeba measures of random plane curves”, Trans. Amer. Math. Soc. 379, 2605–2651 (2026)

work page 2026
[9]

On the expected number of real roots of a system of random polynomial equa- tions

Kostlan, E. “On the expected number of real roots of a system of random polynomial equa- tions”, in Foundations of Computational Mathematics (2002), World Sci. Pub., New Jersey, 149–188

work page 2002
[10]

Positivity in Algebraic Geometry I

Lazarsfeld, R. “Positivity in Algebraic Geometry I”, Springer, (2003). 22 TURGAY BAYRAKTAR AND ALI ULAŞ ÖZGÜR KİŞİSEL

work page 2003
[11]

Chebyshev polynomials

Mason, J. C., Handscomb, D.C. “Chebyshev polynomials”, Chapman & Hall/CRC, (2002)

work page 2002
[12]

‘Òn the number of nodal domains of random spherical harmonics”, American J

Nazarov, F., Sodin, M. ‘Òn the number of nodal domains of random spherical harmonics”, American J. Math., Vol. 131, 1337–1357 (2009)

work page 2009
[13]

Random polynomials of high degree and Levy concentration of measure

Shiffman, B. , Zelditch, S. “Random polynomials of high degree and Levy concentration of measure” Asian J. Math., Vol. 7, No. 4, 627–646 (2003)

work page 2003
[14]

Asymptotics of almost holomorphic sections of ample line bundles on symplectic manifolds

Shiffman, B. Zelditch, S. “Asymptotics of almost holomorphic sections of ample line bundles on symplectic manifolds” J. Reina Angew. Math., Vol. 544, 181–222 (2002)

work page 2002
[15]

Topology of random real hypersurfaces

Welschinger, J-Y. “Topology of random real hypersurfaces” Revista Colombiana Mat., Vol. 49, 139–160 (2015)

work page 2015
[16]

Szegö kernels and a theorem of Tian

Zelditch, S. “Szegö kernels and a theorem of Tian”, Internat. Math. Res. Notices, 317-331 (1998). Faculty of Engineering and Natural Sciences, Sabancı University, Tuzla, Istanbul, Türkiye Email address:tbayraktar@sabanciuniv.edu.tr Department of Mathematics, Middle East Technical University, Ankara, 06800, Türkiye Email address:akisisel@metu.edu.tr

work page 1998

[1] [1]

The Bergman kernel and a theorem of Tian

Catlin, D. “The Bergman kernel and a theorem of Tian” Analysis and geometry in several complex variables, 1–23, Trends Math., Birkhauser (1999)

work page 1999

[2] [2]

On the∂equation in weightedL 2 notms inC 1

Christ, M. “On the∂equation in weightedL 2 notms inC 1” J. Geometric Anal., 1, 193–230 (1991)

work page 1991

[3] [3]

Low degree approximation of random polynomials

Diatta, D. N., Lerario, A. “Low degree approximation of random polynomials” Foundations of Computational Math., Vol. 22, 77–97 (2022)

work page 2022

[4] [4]

How many zeros of a random polynomials are real?

Edelman, A. , Kostlan, E. “How many zeros of a random polynomials are real?” Bull. Amer. Math. Soc. 32, no. 1, 1–37 (1995)

work page 1995

[5] [5]

On the number of connected components of random algebraic hypersurfaces

Fyodorov, Y. V., Lerario, A., Lundberg, E. “On the number of connected components of random algebraic hypersurfaces” J. Geometry and Physics, Vol. 95, 1–20 (2015)

work page 2015

[6] [6]

Betti numbers of random real hypersurfaces and determinants of random symmetric matrices

Gayet, D. , Welschinger, J-Y. “Betti numbers of random real hypersurfaces and determinants of random symmetric matrices”, J. European Matth. Soc., Vol. 18, 732–772 (2016)

work page 2016

[7] [7]

Lower estimates for the expected Betti numbers of random real hypersurfaces

Gayet, D. , Welschinger, J-Y. “Lower estimates for the expected Betti numbers of random real hypersurfaces”, J. London Math. Soc. Vol. 90, 105–120 (2014)

work page 2014

[8] [8]

Amoeba measures of random plane curves

Kişisel, A. U. Ö. , Welschinger, J-Y. “Amoeba measures of random plane curves”, Trans. Amer. Math. Soc. 379, 2605–2651 (2026)

work page 2026

[9] [9]

On the expected number of real roots of a system of random polynomial equa- tions

Kostlan, E. “On the expected number of real roots of a system of random polynomial equa- tions”, in Foundations of Computational Mathematics (2002), World Sci. Pub., New Jersey, 149–188

work page 2002

[10] [10]

Positivity in Algebraic Geometry I

Lazarsfeld, R. “Positivity in Algebraic Geometry I”, Springer, (2003). 22 TURGAY BAYRAKTAR AND ALI ULAŞ ÖZGÜR KİŞİSEL

work page 2003

[11] [11]

Chebyshev polynomials

Mason, J. C., Handscomb, D.C. “Chebyshev polynomials”, Chapman & Hall/CRC, (2002)

work page 2002

[12] [12]

‘Òn the number of nodal domains of random spherical harmonics”, American J

Nazarov, F., Sodin, M. ‘Òn the number of nodal domains of random spherical harmonics”, American J. Math., Vol. 131, 1337–1357 (2009)

work page 2009

[13] [13]

Random polynomials of high degree and Levy concentration of measure

Shiffman, B. , Zelditch, S. “Random polynomials of high degree and Levy concentration of measure” Asian J. Math., Vol. 7, No. 4, 627–646 (2003)

work page 2003

[14] [14]

Asymptotics of almost holomorphic sections of ample line bundles on symplectic manifolds

Shiffman, B. Zelditch, S. “Asymptotics of almost holomorphic sections of ample line bundles on symplectic manifolds” J. Reina Angew. Math., Vol. 544, 181–222 (2002)

work page 2002

[15] [15]

Topology of random real hypersurfaces

Welschinger, J-Y. “Topology of random real hypersurfaces” Revista Colombiana Mat., Vol. 49, 139–160 (2015)

work page 2015

[16] [16]

Szegö kernels and a theorem of Tian

Zelditch, S. “Szegö kernels and a theorem of Tian”, Internat. Math. Res. Notices, 317-331 (1998). Faculty of Engineering and Natural Sciences, Sabancı University, Tuzla, Istanbul, Türkiye Email address:tbayraktar@sabanciuniv.edu.tr Department of Mathematics, Middle East Technical University, Ankara, 06800, Türkiye Email address:akisisel@metu.edu.tr

work page 1998