--- title: Fair Geometric Hitting Set url: https://www.emergentmind.com/topics/fair-geometric-hitting-set type: topic --- # Fair Geometric Hitting Set Searching arXiv for the cited papers and closely related work to ground the article. I’m checking arXiv metadata for the core papers and related geometric hypergraph work. Fair geometric hitting set denotes geometric hitting-set problems in which feasibility is coupled to an explicit fairness requirement. In one formulation, a finite range space $(X,R)$ is equipped with demographic labels on points, and the objective is to select a hitting set whose color composition satisfies demographic parity or prescribed custom ratios while intersecting every range [2507.08758]. In another formulation, fairness is imposed on the coverage pattern itself through a $c$-shallow hitting set: a subset $U \subseteq V$ that meets every hyperedge at least once and at most $c$ times, so that no region is over-covered [2307.12154]. These two lines of work use different fairness semantics, but both convert classical geometric hitting-set feasibility into a constrained balance problem over ranges. ## 1. Formal problem settings The group-fair formulation is defined on a finite range space $(X,R)$ with $\lvert X\rvert = n$ and VC-dimension $d$, where each point $p \in X$ carries a demographic color $c(p)$ in a set $C=\{c_1,\dots,c_k\}$. For each group, $$ X_{c_\ell}=\{\,p\in X \mid c(p)=c_\ell\},\quad \ell=1,\dots,k. $$ A subset $S \subseteq X$ is a fair geometric hitting set when it hits all ranges and simultaneously satisfies one of the prescribed fairness conditions on its demographic composition [2507.08758]. The coverage-fair formulation is stated for a hypergraph $H=(V,E)$. A $c$-shallow hitting set is a vertex subset $U \subseteq V$ such that for every edge $e \in E$, $$ 1 \le |e \cap U| \le c. $$ The lower bound enforces coverage, while the upper bound prevents concentrated selection inside any one region. The source explicitly interprets this as a fairness constraint: each region is hit, but no region is overloaded with too many selected points [2307.12154]. A plausible implication is that “fair geometric hitting set” is not a single canonical optimization problem in the recent literature. Rather, it comprises at least two technically distinct constraint systems: one balancing representation across demographic groups, and another balancing the intensity of coverage across ranges. ## 2. Fairness models In the demographic formulation, the paper introduces two exact group-fairness notions. A subset $S \subseteq X$ satisfies demographic parity (DP) if it preserves the input proportions exactly: $$ \forall\,\ell\in[k]\quad \frac{|S\cap X_{c_\ell}|}{|S|} = \frac{|X_{c_\ell}|}{|X|}. $$ It satisfies custom-ratios fairness (CR) if it matches a target vector $\tau=(\tau_1,\dots,\tau_k)$ with $\sum_\ell \tau_\ell = 1$ exactly: $$ \forall\,\ell\in[k]\quad \frac{|S\cap X_{c_\ell}|}{|S|} = \tau_\ell. $$ When $S$ must also hit all ranges in $R$, these constraints define the corresponding fair geometric hitting set problem [2507.08758]. In the shallow-hitting formulation, fairness is encoded not by demographic balance but by bounded multiplicity of coverage. The selected subset must intersect each hyperedge in at least one and at most $c$ vertices. The accompanying interpretation is that such a set ensures uniform or “fair” representation across regions because no single region receives too many selected points [2307.12154]. These notions are structurally different. DP and CR constrain the global composition of the selected set across demographic groups. The $c$-shallow condition constrains the local interaction between the selected set and every individual range. This suggests that fairness in geometric hitting-set research is best understood as a family of balance constraints, not a single axiom. ## 3. Algorithmic reduction through fair $\varepsilon$-nets The 2025 work reduces fair geometric hitting set to fair $\varepsilon$-net construction [2507.08758]. An $\varepsilon$-net is any $N \subseteq X$ such that every range $R \in R$ with $|R|\ge \varepsilon n$ satisfies $N \cap R \neq \emptyset$. Two algorithms are developed for DP-fair $\varepsilon$-nets. The sampling-based method, FairMonteCarlo$(X,R,\varepsilon,\delta)$, oversamples by an $O(\log k)$ factor so that the random draw is likely to respect each color’s ratio within concentration bounds. Its running time is $O(n+\lambda)\approx O(n)$. With high probability $1-\delta$, it returns a DP-fair $\varepsilon$-net of size $$ O\Bigl(\frac{1}{\varepsilon}\max\{\ln\tfrac{1}{\varepsilon},\,d\ln\tfrac{1}{\varepsilon}\}\,\ln\frac{k}{\delta}\Bigr), $$ which the source describes as only an $O(\ln k)$ overhead over the classical net size [2507.08758]. The discrepancy-based method is deterministic. It repeatedly halves the point set via a low-discrepancy coloring while maintaining exact color ratios. After $$ O\!\Bigl(\log \bigl(n/\!(\tfrac{d}{\varepsilon}\ln(d/\varepsilon))\bigr)\Bigr) $$ iterations, it returns an $\varepsilon$-net of size $$ O\Bigl(\frac{d}{\varepsilon}\ln\tfrac{d}{\varepsilon}\Bigr), $$ matching the best known bound up to constants while preserving demographic parity exactly. Its key ingredients are fair matching of same-color points, conditional-expectation derandomization of random halving, and repeated discrepancy control over ranges [2507.08758]. For fair geometric hitting set itself, the paper uses an LP relaxation with fairness constraints. If $z_p$ are fractional variables, the relaxation minimizes $\sum_{p\in X} z_p$ subject to hitting constraints for all ranges and ratio constraints $$ \sum_{p\in X_{c_\ell}} z_p = \tau_\ell \sum_{p\in X} z_p \quad (\ell=1,\ldots,k), $$ with $0 \le z_p \le 1$. Writing $f=\sum_p z_p$ and $w_p=z_p/f$, every range has weight at least $1/f \equiv \varepsilon$, and the color weights satisfy $\sum_{p\in X_{c_\ell}} w_p = \tau_\ell$ [2507.08758]. Running the sampling-based fair $\varepsilon$-net algorithm on the weighted instance with $\varepsilon = 1/f^*$ yields a hitting set of size $$ O\!\bigl(f^*\ln f^*\ln k\bigr) \subseteq O\!\bigl(\OPT_{\rm FGHS}\,\ln\OPT_{\rm FGHS}\,\ln k\bigr), $$ and therefore an $$ O\bigl(\ln(\OPT)\,\ln k\bigr) $$ approximation for Fair GHS [2507.08758]. ## 4. Shallow hitting sets and polychromatic colorings The 2023 work studies the relation between $c$-shallow hitting sets and polychromatic colorings of hypergraphs [2307.12154]. A polychromatic $k$-coloring is a $k$-coloring of the vertex set such that every hyperedge contains a vertex of all $k$ color classes. The key lemma states that if a hereditary hypergraph family $\mathcal{H}$ satisfies two conditions—closure under induced subhypergraphs, and the existence of a $c$-shallow hitting set in every $m$-uniform member $H_{=m}$ for all $m \ge c$—then its polychromatic parameter obeys $$ m_{\mathcal H}(k)\le c\cdot (k-1)+1. $$ The proof idea is iterative peeling: remove a $c$-shallow hitting set, assign one color, and continue on the remaining induced subhypergraph until the final layer is colored [2307.12154]. This lemma is significant because it converts a bounded-overcoverage statement into a coloring guarantee linear in $k$. The source makes the dependence explicit: if one proves a constant $c$ for the uniform subfamily, then the polychromatic parameter is linear in $k$ by $$ m(k)=O(c\cdot k). $$ Within the fairness interpretation, a smaller shallowity parameter yields stronger balance across the color classes or hitting-set layers, because each layer can be peeled while controlling how heavily any range is represented [2307.12154]. The same paper emphasizes that its contribution is existential and extremal. Explicit polynomial-time algorithms or complexity classifications for finding $c$-shallow hitting sets in the geometric settings under consideration are not developed [2307.12154]. ## 5. Geometric range families and known bounds The geometric families singled out in the hypergraph-based treatment illustrate how fair coverage constraints depend sharply on range geometry [2307.12154]. | Family | Definition or structure | Shallow hitting-set status | |---|---|---| | $\mathcal{B}$ | Bottomless rectangles: axis-parallel regions $\{(x,y): x_0