Chromatic Metric Pairs: Theory and Applications

• Chromatic metric pairs are a family of mathematical frameworks that integrate metric and color structures to study forbidden configurations, distance graphs, and color-discrimination models.
• They provide exponential estimates for chromatic numbers in max-norm settings through combinatorial and density arguments, unifying concepts in Ramsey theory and topological data analysis.
• The theory extends to diverse applications, including stability in colored persistence modules and local Fisher-information metrics that model human color discrimination.

Chromatic metric pairs are studied in several related senses across contemporary geometry, combinatorics, topology, and color science. In one formulation, a pair consists of an ambient metric space and a forbidden metric space, and the basic problem is to color the ambient space with as few colors as possible while avoiding monochromatic isometric copies of the forbidden space. In another, a chromatic metric pair is a metric space together with a coloring of a distinguished subset, so that both geometry and categorical labels enter the comparison problem. A further, more literal use treats nearby color stimuli as metric pairs whose discriminability is governed by a local quadratic form. Across these settings, the common feature is that color information is not added after the metric structure is fixed; it is built into the mathematical object itself (Kupavskii et al., 2020, Draganov et al., 24 Jul 2025).

1. Formal frameworks

The literature uses closely related but non-identical formalizations. In the geometric-Ramsey setting, if $X=(X,\rho_X)$ and $Y=(Y,\rho_Y)$ are metric spaces, a subset $Y'\subset X$ is a copy of $Y$ if there is an isometry $f:Y\to Y'$, and the chromatic number $\chi(X,Y)$ or $\chi(\mathbb X;\mathcal Y)$ is the smallest $k$ such that $X$ can be colored with $k$ colors and no monochromatic copy of $Y=(Y,\rho_Y)$0 appears. In the distance-graph setting, one fixes a metric space $Y=(Y,\rho_Y)$1 and a distance set $Y=(Y,\rho_Y)$2, forms the graph $Y=(Y,\rho_Y)$3 with edges $Y=(Y,\rho_Y)$4, and studies its chromatic number $Y=(Y,\rho_Y)$5. In chromatic topological data analysis, a chromatic metric pair is $Y=(Y,\rho_Y)$6, where $Y=(Y,\rho_Y)$7 is a metric space and $Y=(Y,\rho_Y)$8 colors a subspace $Y=(Y,\rho_Y)$9 (Kloeckner, 2013, Parlier et al., 2014, Draganov et al., 24 Jul 2025).

Framework	Basic object	Associated quantity
Geometric Ramsey theory	$Y'\subset X$0 or $Y'\subset X$1	$Y'\subset X$2, forbidding monochromatic copies
Distance-graph coloring	$Y'\subset X$3 or $Y'\subset X$4	$Y'\subset X$5, forbidding exact distances
Chromatic TDA	$Y'\subset X$6	$Y'\subset X$7 and the six-pack
Color discrimination	local pairwise differences in $Y'\subset X$8	Fisher-information metric $Y'\subset X$9

These frameworks differ in what is treated as primary. The Ramsey formulation emphasizes forbidden finite configurations. The distance-graph formulation emphasizes exact metric shells such as $Y$0. The TDA formulation emphasizes constrained maps and stability under perturbation. The color-discrimination formulation emphasizes infinitesimal pairwise distinguishability. This suggests that “chromatic metric pairs” is best understood as a family of theories organized around joint metric-color structure rather than a single invariant.

2. Max-norm Ramsey theory and universal exponential growth

The most systematic combinatorial theory of chromatic metric pairs concerns $Y$1, where $Y$2 is $Y$3 with

$Y$4

Kupavskii and Sagdeev showed that every finite metric space $Y$5 with at least two points is exponentially $Y$6-Ramsey: there exists $Y$7 such that

$Y$8

Equivalently, for every nontrivial finite forbidden metric space, the chromatic number of the pair $Y$9 grows exponentially in the ambient dimension. A trivial upper bound is always available,

$f:Y\to Y'$0

and the paper recalls the folklore identity $f:Y\to Y'$1 (Kupavskii et al., 2020).

The structural strengthening is the notion of $f:Y\to Y'$2-super-Ramsey. A finite metric space $f:Y\to Y'$3 is $f:Y\to Y'$4-super-Ramsey with parameters $f:Y\to Y'$5 if there are sets $f:Y\to Y'$6 such that

$f:Y\to Y'$7

and every subset of $f:Y\to Y'$8 of size greater than

$f:Y\to Y'$9

contains a copy of $\chi(X,Y)$0. By the pigeonhole principle this implies the exponential lower bound. The paper proves the stronger statement that any finite metric space is $\chi(X,Y)$1-super-Ramsey, and the proof runs through three mechanisms: one-dimensional “batons”, the Frankl–Rödl product theorem for $\chi(X,Y)$2-super-Ramsey spaces, and the classical Fréchet embedding of any finite metric space into a finite-dimensional $\chi(X,Y)$3 space.

For batons, the paper proves a discrete density theorem: $\chi(X,Y)$4 This yields

$\chi(X,Y)$5

For arbitrary real batons $\chi(X,Y)$6, the proof uses a Diophantine-approximation lemma that replaces real increments by an integer model while preserving the relevant partial sums. The paper also gives general upper bounds in terms of the diameter

$\chi(X,Y)$7

and the chain-connectivity threshold $\chi(X,Y)$8: $\chi(X,Y)$9 and, in the rationally scaled integer case,

$\chi(\mathbb X;\mathcal Y)$0

The simplest example is the two-point unit metric $\chi(\mathbb X;\mathcal Y)$1, for which

$\chi(\mathbb X;\mathcal Y)$2

For integer batons $\chi(\mathbb X;\mathcal Y)$3, the lower bound $\chi(\mathbb X;\mathcal Y)$4 and the upper bounds above imply the asymptotic summary

$\chi(\mathbb X;\mathcal Y)$5

For the non-integer three-point baton $\chi(\mathbb X;\mathcal Y)$6, the paper proves

$\chi(\mathbb X;\mathcal Y)$7

while also emphasizing that the correct exponential base is not known in general.

3. Exact asymptotics for batons, products, and infinite forbidden spaces

“Max-norm Ramsey Theory” refines the preceding universal exponential theorem by identifying the exact exponential base for broad classes of one-dimensional forbidden metric spaces. A baton is written

$\chi(\mathbb X;\mathcal Y)$8

where $\chi(\mathbb X;\mathcal Y)$9 has positive entries. For integer batons $k$0, the asymptotic formula is

$k$1

where $k$2 is the supremum of upper densities of subsets $k$3 containing neither a translate $k$4 nor a reflected translate $k$5. For arbitrary batons, if $k$6 is an isomorphism from the additive subgroup generated by the gaps and $k$7, then

$k$8

Thus the exponential base is the reciprocal of an additive-combinatorial density parameter (Frankl et al., 2021).

Several special cases are explicit. If $k$9 are linearly independent over $X$0, then

$X$1

For arithmetic progressions $X$2, Cartesian powers satisfy

$X$3

so the product dimension $X$4 does not change the exponential base. More generally,

$X$5

and if $X$6 is a hyperrectangle, meaning a Cartesian product of several two-point sets, then

$X$7

The theory also reveals strong arithmetic discontinuities. The paper notes

$X$8

but for irrational $X$9 arbitrarily close to $k$0,

$k$1

This shows that the asymptotic base need not vary continuously with the gap vector.

Infinite forbidden metric spaces behave differently from the Euclidean setting. For any infinite $k$2,

$k$3

At the same time, the paper constructs geometric progressions

$k$4

with two distinct properties. For each fixed $k$5, sufficiently small $k$6 gives

$k$7

and there is also a fixed $k$8 such that for any $k$9,

$Y=(Y,\rho_Y)$00

Hence there exists an infinite metric space $Y=(Y,\rho_Y)$01 such that $Y=(Y,\rho_Y)$02 as $Y=(Y,\rho_Y)$03. This is a phenomenon absent in the Euclidean theory described in the same paper.

4. Exact-distance, non-Euclidean, and pseudo-Euclidean variants

A parallel tradition studies exact-distance graphs rather than forbidden finite metric subspaces. Given a metric space $Y=(Y,\rho_Y)$04 and $Y=(Y,\rho_Y)$05, the graph $Y=(Y,\rho_Y)$06 has vertex set $Y=(Y,\rho_Y)$07 and an edge between $Y=(Y,\rho_Y)$08 and $Y=(Y,\rho_Y)$09 exactly when $Y=(Y,\rho_Y)$10. Kloeckner used this framework to show that for translation-invariant metrics on $Y=(Y,\rho_Y)$11 inducing the usual topology, every finite integer and $Y=(Y,\rho_Y)$12 occur as one-distance chromatic numbers: $Y=(Y,\rho_Y)$13 For proper planar metrics, the possibilities are more constrained: $Y=(Y,\rho_Y)$14 and every composite integer $Y=(Y,\rho_Y)$15 belongs to $Y=(Y,\rho_Y)$16. The same paper gives hyperbolic-plane bounds

$Y=(Y,\rho_Y)$17

$Y=(Y,\rho_Y)$18

and

$Y=(Y,\rho_Y)$19

These results frame chromatic metric pairs as exact-distance shell problems rather than copy-avoidance problems (Kloeckner, 2013).

For complete hyperbolic surfaces, the exact-distance formulation becomes genuinely scale-dependent. The $Y=(Y,\rho_Y)$20-chromatic number $Y=(Y,\rho_Y)$21 is the minimum number of colors needed so that points at distance exactly $Y=(Y,\rho_Y)$22 receive different colors. For every complete hyperbolic surface $Y=(Y,\rho_Y)$23,

$Y=(Y,\rho_Y)$24

and there is a family $Y=(Y,\rho_Y)$25 with

$Y=(Y,\rho_Y)$26

For closed genus-$Y=(Y,\rho_Y)$27 hyperbolic surfaces, the full chromatic number

$Y=(Y,\rho_Y)$28

satisfies

$Y=(Y,\rho_Y)$29

while some genus-$Y=(Y,\rho_Y)$30 surfaces satisfy

$Y=(Y,\rho_Y)$31

The proofs use maximal separated sets, hyperbolic area growth, thick–thin decomposition, collars around short geodesics, and geometric embeddings of complete graphs (Parlier et al., 2014).

Pseudo-Euclidean or spacetime variants replace positive-definite distance by an indefinite quadratic form. In the signature-$Y=(Y,\rho_Y)$32 setting, every finite coloring of $Y=(Y,\rho_Y)$33 contains a monochromatic pair with

$Y=(Y,\rho_Y)$34

Consequently,

$Y=(Y,\rho_Y)$35

and by embedding,

$Y=(Y,\rho_Y)$36

In $Y=(Y,\rho_Y)$37, if $Y=(Y,\rho_Y)$38, every $Y=(Y,\rho_Y)$39-coloring contains a monochromatic pair with

$Y=(Y,\rho_Y)$40

The paper also proves a density theorem in $Y=(Y,\rho_Y)$41, showing that pseudo-Euclidean chromatic pair phenomena can be approached through Cayley graphs and Fourier-analytic ratio bounds rather than only through exact-distance graph coloring (Davies, 2023).

5. Chromatic metric pairs in topological data analysis

In chromatic topological data analysis, a chromatic metric pair is

$Y=(Y,\rho_Y)$42

where $Y=(Y,\rho_Y)$43 is a metric space and $Y=(Y,\rho_Y)$44 is a colored subspace. A metric pair $Y=(Y,\rho_Y)$45 is recovered as the constant-color case $Y=(Y,\rho_Y)$46. The theory introduces constraint sets $Y=(Y,\rho_Y)$47: a map $Y=(Y,\rho_Y)$48 is $Y=(Y,\rho_Y)$49-constrained if

$Y=(Y,\rho_Y)$50

and $Y=(Y,\rho_Y)$51-constrained if this holds for every $Y=(Y,\rho_Y)$52. The trivial constraint set is $Y=(Y,\rho_Y)$53, while the discrete constraint set is $Y=(Y,\rho_Y)$54, which forces exact color preservation (Draganov et al., 24 Jul 2025).

The central comparison invariant is the $Y=(Y,\rho_Y)$55-constrained Gromov–Hausdorff distance

$Y=(Y,\rho_Y)$56

where the infimum is taken over $Y=(Y,\rho_Y)$57-constrained maps $Y=(Y,\rho_Y)$58 and $Y=(Y,\rho_Y)$59. The paper proves correspondence and common-embedding characterizations analogous to classical Gromov–Hausdorff theory, and shows that for metric spaces $Y=(Y,\rho_Y)$60,

$Y=(Y,\rho_Y)$61

For metric pairs $Y=(Y,\rho_Y)$62, $Y=(Y,\rho_Y)$63, the induced distance satisfies

$Y=(Y,\rho_Y)$64

The same paper introduces the six-pack, a collection of six persistence diagrams derived from the inclusion

$Y=(Y,\rho_Y)$65

between two color-pattern subfiltrations of the ambient Čech filtration. The six associated persistence modules are the domain, codomain, image, kernel, cokernel, and relative module. If $Y=(Y,\rho_Y)$66 and $Y=(Y,\rho_Y)$67 are finite, $Y=(Y,\rho_Y)$68 are finite-dimensional simplicial complexes on $Y=(Y,\rho_Y)$69, and

$Y=(Y,\rho_Y)$70

then for each of the six corresponding persistence modules $Y=(Y,\rho_Y)$71,

$Y=(Y,\rho_Y)$72

Restricting to ordinary metric pairs yields the ambient Čech stability bound

$Y=(Y,\rho_Y)$73

This makes the chromatic metric pair formalism a direct extension of both metric-pair stability theory and chromatic point-set TDA.

6. Local chromatic metrics in color discrimination

A distinct use of the subject concerns chromatic metric pairs in the literal local sense of nearby color comparisons. The paper “A Metric for Three-Dimensional Color Discrimination Derived from V1 Population Fisher Information” derives a Riemannian metric on three-dimensional color space from the Fisher information of neural population codes in the visual pathway. For nearby colors differing by $Y=(Y,\rho_Y)$74, the infinitesimal discrimination distance is

$Y=(Y,\rho_Y)$75

and the paper interprets the Fisher information matrix as the local perceptual metric. The color space coordinates are $Y=(Y,\rho_Y)$76, where $Y=(Y,\rho_Y)$77 are CIE chromaticity coordinates and $Y=(Y,\rho_Y)$78 is luminance (Menke, 25 Mar 2026).

The construction proceeds from CIE tristimulus values to cone excitations, then to von Kries adaptation, retinal opponent coordinates, and a three-population V1 encoding model. In opponent coordinates $Y=(Y,\rho_Y)$79, the chromatic Fisher metric is

$Y=(Y,\rho_Y)$80

with pathway-specific terms for parvocellular, koniocellular, and non-cardinal populations. The chromaticity metric is obtained by pullback,

$Y=(Y,\rho_Y)$81

the luminance term is

$Y=(Y,\rho_Y)$82

and the full $Y=(Y,\rho_Y)$83 metric is block diagonal: $Y=(Y,\rho_Y)$84 At threshold, nearby just-detectable displacements satisfy a quadratic-form equation, so the inverse metric determines the threshold ellipses in chromaticity and threshold ellipsoids in full $Y=(Y,\rho_Y)$85-space.

The model has 17 fitted parameters and is jointly fitted to four datasets: MacAdam’s chromaticity ellipses, Koenderink et al.’s three-dimensional ellipsoids, Wright’s wavelength discrimination function, and Huang et al.’s threshold color-difference ellipses, covering 96 independently measured discrimination conditions. The reported STRESS values are 23.9 on MacAdam, 20.8 on Koenderink et al., 30.1 on Wright, and 30.8 on Huang et al. The paper emphasizes that the construction is local rather than global: for separated color pairs one would need path integration of the line element, whereas the empirical validation is carried out through threshold ellipses and ellipsoids.

Taken together, these literatures show that chromatic metric pairs are not a single isolated definition but a recurrent mathematical pattern. In combinatorics they encode forbidden monochromatic isometric copies or exact-distance shells; in TDA they encode colored subspaces and constrained correspondences; in perceptual geometry they encode local pairwise discriminability through a metric tensor. The unifying theme is that metric structure and chromatic structure are treated as a coupled object rather than as independent data.