Topological triviality and link-constancy in deformations of inner Khovanskii non-degenerate maps

Basic Information

• Paper ID: 2504.18816
• Title: Topological triviality and link-constancy in deformations of inner Khovanskii non-degenerate maps
• Authors: Julian D. Espinel Leal, Eder L. Sanchez Quiceno
• Classification: math.AG (Algebraic Geometry), involving singularity theory
• Publication Date: October 23, 2025 (arXiv v2)
• Paper Link: https://arxiv.org/abs/2504.18816

Abstract

This paper introduces the concept of Inner Khovanskii Non-Degeneracy (IKND) for real polynomial maps and mixed polynomial maps $f=(f^1,\dots,f^p)$ (satisfying $f(0)=0$), generalizing the non-degeneracy condition introduced by Wall in 1999 for complex polynomial functions. The authors prove that IKND is a sufficient condition for ensuring that the link of the singularity at the origin is smooth and well-defined. Subsequently, they study one-parameter deformations $F(x,\varepsilon)=f(x)+\theta(x,\varepsilon)$ of IKND maps and prove that under appropriate conditions, the deformations possess link-constancy, meaning the ambient isotopy type of the link remains invariant along the deformation. Furthermore, by adopting a strong version of non-degeneracy (SIKND), results on topological triviality are obtained. Finally, inspired by work of Araújo dos Santos et al., the paper explores deformations of mixed polynomial functions in two variables and provides a characterization of links in a special class of deformations.

Research Background and Motivation

Core Problem

A key question in complex (or real) analytic singularity theory is: given a deformation $F(x,\varepsilon): K^n \times K \to K^p$ (where $K=\mathbb{C}$ or $\mathbb{R}$), do the map families $F_\varepsilon(x):=F(x,\varepsilon)$ preserve local equivalence (equisingularity or triviality) for sufficiently small values of $\varepsilon$?

Research Significance

1. Generalization of Classical Results: The foundational results of Lê and Ramanujam show that when $n\neq 3$, complex analytic function deformations with constant Milnor number are V-trivial. Timourian further proved topological triviality. Wall in 1999 studied deformations of complex analytic functions and proved topological triviality under the assumption of inner non-degeneracy (INND), even in the difficult case $n=3$.
2. Natural Extension to Mixed Polynomials: Mixed polynomial functions (complex-valued polynomials in complex variables and their conjugates) are equivalent to real polynomial maps $\mathbb{R}^{2n}\to\mathbb{R}^2$, with complex functions as special cases. Studying mixed polynomials via Newton polyhedra has significant advantages and allows construction of families of isolated real singularities.
3. Bridging Complex and Real Theories: Existing research on equisingularity of deformations is primarily concentrated in the complex setting. This paper bridges the gap between complex and real singularity theories through a unified non-degeneracy framework.

Limitations of Existing Methods

• Wall's INND condition applies only to complex polynomial functions
• Araújo dos Santos et al. generalized inner non-degeneracy to two-variable mixed polynomials but lacked a more general framework
• Equisingularity theory for real polynomial maps and higher-dimensional mixed maps remains incomplete
• Existing non-degeneracy conditions (such as KND) may be too restrictive and inapplicable to certain important examples

Core Contributions

1. Introduction of (S)IKND Concepts: Defines Inner Khovanskii Non-Degeneracy (IKND) and its strong version (SIKND) for real polynomial maps and mixed polynomial maps, based on sets of C-face diagrams, generalizing Wall's INND and classical KND.
2. Smoothness of Links: Proves that IKND is a sufficient condition for ensuring that the singularity link is smooth and well-defined (Theorem 3.3).
3. Non-Accumulation of Critical Points: Establishes sufficient conditions ensuring that (weak) critical points of deformations do not accumulate, which is crucial for subsequent study of link-constancy and topological triviality.
4. Link-Constancy Results:
   • Theorem 4.7: For KND maps $f_D$ and IKND maps $f$, under appropriate conditions deformations are link-constant; if $f$ is SIKND then topologically trivial
   • Theorem 4.11: Deformations of SWH/SRWH maps are link-constant under appropriate conditions
   • Theorems 4.14 and 4.17: Link-constancy of deformations of two-variable mixed polynomial functions
5. Complete Characterization for Two-Variable Mixed Functions: Generalizes the classical result that the link isotopy type of a convenient non-degenerate complex polynomial function is determined by terms on the Newton boundary (Corollary 4.15).
6. Rich Classes of Examples: Proves that convenient KND maps, semi-weighted homogeneous maps (SWH), semi-radial weighted homogeneous mixed maps (SRWH), and three-variable non-convenient KND functions are all examples of IKND.

Methodology Details

Core Concepts

C-face Diagram

Given a set of homogeneous linear functions $J$, where each $\ell$ satisfies $\ell(\nu)>0$ for all $\nu\in\mathbb{R}^n_{>0}$. Define the minimal function $\ell_J(\nu)=\min_{\ell\in J}\ell(\nu)$. The C-face diagram is: $D(J) = \{\nu\in\mathbb{R}^n_{\geq 0} \mid \ell_J(\nu)=1\}$

An inner face of the C-face diagram is a face not contained in any proper coordinate subspace.

IKND Definition (Definition 2.1)

A map $f=(f^1,\ldots,f^p)$ is called Inner Khovanskii Non-Degenerate (IKND) if there exist C-face diagrams $D_1,\ldots,D_p$ satisfying:

(i) $\text{supp}(f^i) \subseteq D_i + \mathbb{R}^n_{\geq 0}$, i.e., the support of $f^i$ does not lie below $D_i$

(ii) For each inner face $\Delta$ of $D=D_1+\cdots+D_p$ and each non-empty subset $I\subseteq[n]$: $\Delta \cap \mathbb{R}^I \neq \emptyset \Rightarrow \Sigma(f_\Delta) \cap V(f_\Delta) \cap (K^*)^I = \emptyset$

where $f_\Delta=(f^1_{\Delta_1},\ldots,f^p_{\Delta_p})$ is the face system corresponding to $\Delta=\Delta_1+\cdots+\Delta_p$.

Strong Version (SIKND): Strengthens condition (ii) to $\Delta \cap \mathbb{R}^I \neq \emptyset \Rightarrow \Sigma(f_\Delta) \cap (K^*)^I = \emptyset$

Key Properties

Non-Accumulation of Critical Points (Definition 3.1): A deformation has (weak) non-accumulation of critical points if there exists $\epsilon>0$ such that for all $\varepsilon\in K$, $\Sigma(F_\varepsilon)\cap B_\epsilon(0) \subseteq \{0\}$ (or only requiring $\Sigma(F_\varepsilon)\cap V(F_\varepsilon)\cap B_\epsilon(0)\subseteq\{0\}$).

$\rho$-Uniform Milnor Radius (Definition 4.3): There exists $\epsilon_0>0$ such that for all $0<\epsilon\leq\epsilon_0$ and $\varepsilon\in K$, $V(F_\varepsilon)$ intersects $\rho^{-1}(\epsilon)$ transversely.

Link-Constancy (Definition 4.1): For all $\varepsilon\in K$, the link $L_{F_\varepsilon}$ is ambient isotopic to $L_{F_0}$.

Main Theorems

Theorem 3.3 (Non-Accumulation of Critical Points)

If there exist C-face diagrams $D_1,\ldots,D_p$ satisfying:

• $f$ is IKND with respect to $\{D_j\}$
• For each $j$ and $\varepsilon$, $d(w;F^j_\varepsilon-f^j)>d(w;D_j)$ for all $w\in W(D_j)$

then the deformation has weak non-accumulation of critical points. If $f$ is SIKND, then it has non-accumulation of critical points.

Proof Strategy:

1. By curve selection lemma, assume existence of accumulating critical point sequence
2. Construct weight vectors $q,q',w$ and use Lemma 3.2 to establish inner face relations
3. Via weighted degree analysis (Claim 3.4), establish gradient relations for face functions
4. Derive contradiction: some point belongs to $\Sigma(f_\Theta)\cap(K^*)^{I_\Theta}$, violating IKND condition

Theorem 4.7 (Link-Constancy and Topological Triviality)

If there exist C-face diagrams $D_1,\ldots,D_p$ satisfying:

• $f_D$ is KND
• $f$ is IKND with respect to $\{D_j\}$
• The deformation is above $D_j$
• Technical condition (iv): for inner faces and index sets satisfying specific conditions, there exists a minimal weight component

then the deformation is link-constant. If $f$ is SIKND, then it is topologically trivial.

Proof Core:

1. By Proposition 4.4, link-constancy is equivalent to proving existence of $\rho_E$-uniform Milnor radius
2. Proof by contradiction: assume no uniform radius exists, apply curve selection lemma
3. Analyze two cases:
   • Case I: $I'\cap I_\Theta\neq\emptyset$, derive contradiction using Euler identity
   • Case II: $I'\cap I_\Theta=\emptyset$, derive contradiction using condition (iv) and KND property
4. Topological triviality follows from Proposition 4.6

Theorem 4.11 (SWH/SRWH Maps)

For deformations of semi-weighted homogeneous maps, under appropriate conditions:

• Link is constant in some neighborhood
• If inequality is strict, then globally link-constant
• If $\Sigma(f_w)=\{0\}$ and inequality is strict, then topologically trivial

Proof Method:

1. Use polar coordinate transformation $\pi(s,r)=(r^{w_1}s_1,\ldots,r^{w_n}s_n)$
2. Transform problem to zero set on $\rho^{-1}(1)\times\mathbb{R}_{\geq 0}$
3. Apply implicit function theorem to construct isotopy
4. Distinguish two cases: $I=\emptyset$ and $I\neq\emptyset$

Theorem 4.17 (Two-Variable Mixed Functions)

For deformations of two-variable mixed polynomial functions, under IKND and technical conditions, the deformation is link-constant; if SIKND then topologically trivial.

Key Technique:

1. Use Claim 2.11 to establish weight vector relations
2. Analyze two cases of $I'\cap I_\Theta$
3. In Case II, derive contradiction through divisibility arguments and monomial analysis

Technical Innovations

1. C-face Diagram Framework: Unlike classical KND using only Newton boundary, IKND allows more flexible C-face diagram choices, accommodating broader classes of maps (see example in Remark 2.2).
2. Inner Face Condition: Requires non-degeneracy only on inner faces rather than all faces, representing essential generalization of Wall's INND concept.
3. Weight Vector Construction: The construction technique in Lemma 3.2, starting from projected weight vector $\pi_I(q)$ and progressively refining to inner face weight vector $w$, is the key tool in the proof.
4. Correspondence between Mixed and Real Maps: Lemma 2.10 establishes relationship between (S)IKND properties of mixed polynomials and corresponding real maps, though the converse does not hold (Remark 2.12).
5. Uniform Radius Method: Proposition 4.4 transforms link-constancy into existence of $\rho$-uniform Milnor radius, providing unified proof framework.

Experimental Setup

Theoretical Verification through Examples

As a pure mathematics theory paper, this work does not involve traditional experiments but rather verifies theoretical results through constructive examples:

Example 2.9 (SIKND but not SKND)

$f(x_1,x_2,x_3) = x_1^{12} + x_1x_2^4x_3 + (x_2^3-x_3^2)^2$

• Weight vector $w=(2,2,3)$ corresponds to face function $f_w=(x_2^3-x_3^2)^2$ which is singular at $(1,1,1)$, hence not SKND
• However, by verifying that the partial derivative system has only trivial solutions, we prove $\Sigma(f)=\{0\}$, hence is SIKND

Example 4.13 (Weighted Homogeneous Function)

$f(x_1,x_2,x_3) = x_1^{12} + x_1x_2^4x_3 + (x_2^3-x_3^2)^2$ Weight type $(1,2,3;12)$, applying Theorem 4.11 yields link-constancy and topological triviality.

Example 4.16 (Two-Variable Mixed Function)

$f(x) = x_1^6 + \bar{x}_2x_1^4 + x_1\bar{x}_2^3 + x_2^6$

• Construct C-face diagram $D(J_1)$ defined by linear functions $\ell_{1,1}(\nu)=\langle(\frac{2}{11},\frac{3}{11}),\nu\rangle$ and $\ell_{1,2}(\nu)=\langle(\frac{1}{4},\frac{1}{4}),\nu\rangle$
• $D(J_1)$ allows more deformations than Newton boundary $\Gamma(f)$ (such as $\theta(x,\varepsilon)=\varepsilon x_2^5$)
• Link is characterized by nested link $L([L_1],[L_2]')$

Example 4.19 (Non-nice Function)

$f(x) = x_1^4 + (x_2^2-\bar{x}_2^2)x_1\bar{x}_1 + (i+1)x_2\bar{x}_2(x_1^2+\bar{x}_1^2) + x_2^6$

• Detailed computation (in appendix) verifies IKND but does not satisfy nice condition
• $V(f_{\Delta_3})\cap(\mathbb{C}^*)^2\neq\emptyset$ where $\Delta_3$ is a vertex
• Demonstrates advantage of Theorem 4.17 over Theorem 4.14

Verification Methods

1. Algebraic Verification: Solve critical point equations to verify $\Sigma(f_\Delta)\cap V(f_\Delta)\cap(K^*)^I=\emptyset$
2. Weighted Degree Analysis: Compute face functions and weighted degrees under different weight vectors
3. Counterexample Construction: Show certain real IKND maps do not arise from mixed IKND maps (Remark 2.12)

Experimental Results

Main Theoretical Results

1. Breadth of IKND Class (Propositions 2.4, 2.8, 2.17):
   • Convenient KND maps ⊂ IKND
   • SWH maps ⊂ IKND; if $\Sigma(f_w)=\{0\}$ then ⊂ SIKND
   • Three-variable KND functions satisfying Lemma 2.15 conditions ⊂ IKND
2. Relationship with Existing Concepts (Proposition 2.14):
   • For two-variable mixed functions, (S)IND ⟺ (S)IKND
   • Generalizes definition of Araújo dos Santos et al.
3. Hierarchy of Deformation Properties:

   SIKND + condition (ii) 
     ⟹ non-accumulation of critical points 
     ⟹ link well-defined
   
   SIKND + condition (ii) + uniform radius
     ⟹ link-constancy
   
   SIKND + condition (ii) + uniform radius + (n-p≤2 or other)
     ⟹ topological triviality
   

4. Link Characterization (Corollary 4.15):
   • For nice IKND two-variable mixed functions, link isotopy type is completely determined by $f_D$
   • Link structure is nested link $L([L_1,\ldots,L_{N-1}],[L_N]')$
   • Sublinks constructed from functions $f_{\Delta(w_i;D)}$

Results for Specific Examples

Example 4.10 (Three-Variable Function)

$f(x_1,x_2,x_3) = (x_1^6+x_1x_2^5+x_2x_3^6)(x_1^2+x_2^2+x_3^2)$

• $\text{Inc}(f)=\{2,3\}$
• When $v_2\geq 9, v_3>46/5$, $D_v(f)$ satisfies conditions of Theorem 4.7
• Any deformation above $D_v(f)$ is link-constant

Link Structure of Example 4.16

• $L_1 = \{0\}\times\{0\}\times S^1$
• $L_2$ splits into two branches: $\{0\}\times S^1\times\{0\}$ and trefoil knot given by roots $e^{-\frac{3}{2}it_1},-e^{-\frac{3}{2}it_1}$

Key Findings

1. Non-Uniqueness: C-face diagram choice is not unique (Remark 2.2), but certain choices allow broader deformation classes.
2. Role of Nice Condition: Theorem 4.14 requires nice condition to apply implicit function theorem; Theorem 4.17 avoids this limitation through different techniques.
3. Dimension Restriction: Proposition 4.5 requires $n-p\leq 2$ (real case) or $n-p\leq 1$ (mixed case); when satisfied, non-accumulation of critical points automatically implies topological triviality.
4. Difference between Mixed and Real: Lemma 2.10 and Remark 2.12 show that realization of mixed IKND maps is real IKND, but the converse does not hold.

Related Work

Classical Singularity Theory

1. Lê-Ramanujam (1976): V-triviality of $\mu$-constant deformations ($n\neq 3$)
2. Timourian (1977): Topological triviality of $\mu$-constant deformations ($n\neq 3$)
3. Wall (1999): Topological triviality of complex functions under INND condition, including case $n=3$
4. King (1980/81): Relationship between non-accumulation of critical points and $\mu$-constancy

Non-Degeneracy Conditions

1. Kouchnirenko (1976): Newton non-degeneracy and Milnor number
2. Khovanskii (1977): KND condition
3. Oka (2010): Non-degeneracy for mixed functions
4. Bivià-Ausina (2007): Mixed Newton number and ICIS

Mixed Polynomial Research

1. Oka (2010): Newton polyhedron definition for mixed functions
2. Araújo dos Santos et al. (2024): Inner non-degeneracy for two-variable mixed functions
3. Bode (2019, 2025): Construction of real algebraic links
4. Eyral-Oka (2017, 2022): Geometry of non-isolated singularities

Deformation Theory

1. Damon (1989): Topological invariants of $\mu$-constant deformations of ICIS
2. Gaffney (1992): Integral closure of modules and Whitney equisingularity
3. Bekka (2015): Uniform radius and equisingularity
4. Nguyen (2022): Uniform stable radius for non-degenerate ICIS

Positioning of This Work

The (S)IKND concept in this paper:

• Generalizes: Wall's INND (complex functions) → IKND (real/mixed maps)
• Generalizes: Araújo dos Santos et al.'s IND (two variables) → IKND (arbitrary dimension)
• Generalizes: KND → IKND (allows non-convenient cases and more flexible C-face diagrams)
• Unifies: Provides unified framework for complex and real singularity theories

Conclusions and Discussion

Main Conclusions

1. (S)IKND is an Effective Non-Degeneracy Condition:
   • Ensures link is smooth and well-defined
   • Encompasses multiple important map classes (KND, SWH, SRWH)
   • More general than SKND (Example 2.9)
2. Equisingularity of Deformations:
   • Theorem 3.3 provides sufficient conditions for (weak) non-accumulation of critical points
   • Theorems 4.7, 4.11, 4.17 establish link-constancy and topological triviality
   • Uniform Milnor radius is effective tool for proving link-constancy
3. Complete Theory for Two-Variable Mixed Functions:
   • Generalizes classical results to mixed setting
   • Link isotopy type determined by terms on C-face diagram (Corollary 4.15)
   • Theorem 4.17 avoids nice condition limitation

Limitations

1. C-face Diagram Construction:
   • Finding appropriate C-face diagrams $D_1,\ldots,D_p$ for given map may require techniques
   • Geometric meaning of condition (iv) in Theorem 4.7 not sufficiently intuitive
   • General algorithm for constructing maximal C-face diagram $\Gamma_{\text{inn}}(f)$ not provided
2. Dimension Restrictions:
   • Proposition 4.5 has restrictions on $n-p$
   • Topological triviality in high codimension requires additional conditions (such as uniform radius)
3. Nice Condition:
   • Theorem 4.14 requires nice condition
   • Although Theorem 4.17 avoids this, proof is more complex
   • Geometric meaning and verifiability of nice condition warrant further research
4. Computational Complexity:
   • Verifying IKND condition requires checking all inner faces
   • For complex maps, number of faces may be large
   • Lack of effective algorithmic implementation
5. Case $n=3$:
   • Although Theorems 4.7 etc. apply to $n=3$, Proposition 4.5 does not
   • Topological triviality in three dimensions still requires more refined analysis

Future Directions

1. Algorithmic Aspects:
   • Develop computer-aided tools for verifying IKND condition
   • Algorithm for constructing optimal C-face diagrams
   • Implementation in symbolic computation systems
2. Theoretical Generalizations:
   • Extend to analytic maps
   • Study non-isolated singularity case
   • Explore connections with D-module theory
3. Geometric Characterization:
   • Deeper understanding of geometric meaning of condition (iv)
   • Study topological properties of $\Gamma_{\text{inn}}(f)$
   • Compute topological invariants of links
4. Applications:
   • Applications in real algebraic geometry
   • Connections with knot theory (such as Example 4.16)
   • Applications in robotics and computational geometry
5. Higher-Dimensional Generalizations:
   • Extend to case $p>2$
   • Study stratified link structures
   • Connections with higher-dimensional knot theory

In-Depth Evaluation

Strengths

1. Strong Theoretical Innovation:
   • (S)IKND concept represents substantial generalization of classical non-degeneracy conditions
   • C-face diagram framework provides more flexible tool than Newton boundary
   • Successfully unifies treatment of real and mixed maps, bridging complex and real theories
2. Technical Rigor:
   • Proofs are detailed and complete; weight vector construction technique in Lemma 3.2 is elegant
   • Multiple theorems provide results at different levels (non-accumulation of critical points → link-constancy → topological triviality)
   • Clear distinction between IKND and SIKND, precisely characterizing role of different conditions
3. Rich Examples:
   • Multiple constructive examples verify theory
   • Example 2.9 shows SIKND strictly weaker than SKND
   • Example 4.19 demonstrates advantage of Theorem 4.17 over 4.14
   • Example 4.10 provides concrete C-face diagram construction
4. Complete Results:
   • Provides complete link characterization for two-variable mixed functions (Corollary 4.15)
   • Establishes precise relationships between IKND and existing concepts (IND, KND)
   • Multiple theorems provide results under different hypotheses
5. Clear Exposition:
   • Well-organized structure, progressing from definitions to applications
   • Remarks and examples aid understanding
   • Technical details (such as Claim 3.4) stated separately

Weaknesses

1. Limited Computability:
   • Lacks effective algorithm for verifying IKND
   • C-face diagram selection depends on experience and technique
   • Manual verification of inner face conditions difficult for complex maps
2. Insufficient Geometric Intuition:
   • Geometric meaning of condition (iv) in Theorem 4.7 not sufficiently clear
   • Relationship between C-face diagrams and Newton polyhedra could be explored more deeply
   • Link structure visualization could be enhanced
3. Limited Application Scope:
   • Main results concentrated on polynomial maps
   • Extension to non-polynomial analytic maps not addressed
   • Discussion of practical applications relatively limited
4. Technical Limitations:
   • Some theorems require additional technical conditions (such as nice, condition (iv))
   • Dimension restrictions (Proposition 4.5) limit applicability
   • Verification of uniform radius existence may be difficult
5. Literature Review:
   • While main related works cited, connections to certain fields (such as computational algebraic geometry) insufficiently discussed
   • Comparison with recent symbolic computation methods lacking

Impact

1. Theoretical Contribution:
   • Provides important tools for real singularity theory
   • Advances development of mixed polynomial singularity theory
   • May inspire research on other non-degeneracy conditions
2. Potential Applications:
   • Link classification in real algebraic geometry
   • Singularity avoidance in robotics
   • Algorithm design in computational topology
3. Reproducibility:
   • Theoretical results rigorously proved, strong reproducibility
   • Examples detailed, usable as verification tools
   • Code implementation lacking
4. Subsequent Research:
   • Establishes foundation for research on higher-dimensional mixed maps
   • Inspires development of algorithms and computational tools
   • May generate cross-disciplinary connections (topology, geometry)

Applicable Scenarios

1. Theoretical Research:
   • Study singularities of real/mixed polynomial maps
   • Classify singularity links in specific dimensions
   • Construct maps with specific topological properties
2. Computational Mathematics:
   • Singularity analysis modules in symbolic computation systems
   • Link invariant computation
   • Deformation space parameterization
3. Applied Mathematics:
   • Singularity analysis in robot motion planning
   • Critical point study in optimization problems
   • Topological phase transitions in phase transition theory
4. Education:
   • Advanced courses in singularity theory
   • Specialized topics in algebraic geometry
   • Link theory in topology courses

Selected References

2 R. N. Araújo dos Santos, B. Bode, and E. L. Sanchez Quiceno, Links of Singularities of Inner Non-degenerate Mixed Functions, Bull. Braz. Math. Soc. (N.S.) 55 (2024), no. 3, Paper No. 34.

18 A. G. Khovanskii, Newton polyhedra, and toroidal varieties, Funkcional. Anal. i Priložen. 11 (1977), no. 4, 56–64.

21 D. T. Lê and C. P. Ramanujam, The invariance of Milnor's number implies the invariance of the topological type, Amer. J. Math. 98 (1976), no. 1, 67–78.

29 M. Oka, Non-degenerate mixed functions, Kodai Math. J. 33 (2010), no. 1, 1–62.

33 C. T. C. Wall, Newton polytopes and non-degeneracy, J. Reine Angew. Math. 509 (1999), 1–19.

Overall Assessment: This is a high-quality pure mathematics paper making substantial contributions to singularity theory. The introduction of (S)IKND concepts and their application in deformation theory provide powerful tools for studying real and mixed polynomial maps. The theory is rigorous and results are complete. While further development is needed in algorithmic and application aspects, the paper establishes solid foundation for subsequent research. Particularly noteworthy is the successful generalization of classical results from complex singularity theory to more general real and mixed settings, demonstrating profound mathematical insight.