--- title: Homological Percolation Dynamics url: https://www.emergentmind.com/topics/homological-percolation type: topic --- # Homological Percolation Dynamics Searching arXiv for recent and foundational papers on homological percolation and closely related topological percolation models. Homological percolation is a higher-dimensional analogue of classical percolation in which the relevant macroscopic objects are not only connected components, but also homology classes that remain nontrivial in an ambient space. In the modern formulation developed for random subsets of the flat torus \(T^d\), one studies a filtration \(\{X_t\}\) and the inclusion-induced maps \(H_k(X_t)\to H_k(T^d)\); the resulting nonzero image classes are the “giant” or “essential” \(k\)-cycles that define the percolative transition in degree \(k\) [2005.14011]. Across continuum, lattice, and random-field models, the subject links percolation theory, algebraic topology, random geometric complexes, and topological data analysis, with a recurring empirical theme that the onset of giant cycles is closely related to zeros of the expected Euler characteristic curve [1910.10146]. ## 1. Conceptual framework and formal definitions Classical percolation is fundamentally a \(0\)-dimensional phenomenon: the emergence of a giant connected component is an \(H_0\) event. Homological percolation extends this viewpoint to higher dimensions by asking when a random space contains macroscopic cycles that represent ambient homology classes rather than merely local topological fluctuations [2005.14011]. A standard setting is a random filtration \(\{X_t\}\) of subsets of a compact ambient space \(M\), with \[ X_s \subset X_t \subset M \qquad (s \lambda_{1,k}\), then all giant \(k\)-cycles are exponentially likely. The lowest nontrivial dimension is exact: \[ \lambda_{0,1} = \lambda_{1,1} = \lambda_c. \] Thus giant \(1\)-cycles appear exactly at the classical occupied percolation threshold [2005.14011]. At the opposite end, \[ \lambda_{0,d-1} \le \lambda_{1,d-1} \le \bar \lambda_c, \] so giant \((d-1)\)-cycles are controlled from above by the vacancy threshold [2005.14011]. The threshold hierarchy is monotone: \[ \lambda_{\theta,k} \le \lambda_{\theta,k+1},\qquad \theta=0,1, \] which formalizes the principle that higher-dimensional macroscopic structures emerge no earlier than lower-dimensional ones [2005.14011]. A crucial structural identity is the toroidal duality formula \[ \beta_k(T^d) = \rank(i_k) + \rank(\bar i_{d-k}), \] which yields the complementarity \[ A_k = (\bar E_{d-k})^c. \] This occupied-vacant duality ties giant occupied \(k\)-cycles to giant vacant \((d-k)\)-cycles and underlies the relationship between lower- and upper-dimensional thresholds [2005.14011]. The same ambient-essential notion reappears in finite-volume toroidal lattice models. In plaquette and permutohedral models on \(\mathbb T^d\), an \(i\)-cycle is called giant if its class remains nontrivial in \(H_i(\mathbb T^d;\mathbb F)\), and the sharp-threshold theory concerns the jump from \(\operatorname{rank}\phi_*=0\) to \(\operatorname{rank}\phi_*=\binom di\) [2011.11903]. This places the torus at the center of the subject as the natural ambient space for defining macroscopic cycles. ## 3. Lattice models, duality, and exact threshold phenomena Finite toroidal lattice models provide a rigorous parallel to continuum homological percolation. In plaquette percolation, the ambient complex is the cubical decomposition of \[ \mathbb T^d_N=\mathbb Z^d/(N\mathbb Z)^d, \] and the random subcomplex is formed by taking the full \((i-1)\)-skeleton and then including each \(i\)-face independently with probability \(p\). In the permutohedral model, the ambient torus carries the tiling induced by the Voronoi cells of \(\mathcal A_d^*\), and each \(d\)-dimensional permutohedron is included independently with probability \(p\) together with all its faces [2011.11903]. For plaquettes, the relevant events are \[ A^{\square}=\{\operatorname{rank}\phi_*>0\}, \qquad S^{\square}=\{\operatorname{rank}\phi_*=\binom di\}, \] while for permutohedra, for each \(i\), \[ A_i^{\hexagon}=\{\operatorname{rank}\varphi_{i*}>0\}, \qquad S_i^{\hexagon}=\{\operatorname{rank}\varphi_{i*}=\binom di\}. \] These encode, respectively, the existence of at least one giant cycle and the spanning of all ambient homology in degree \(i\) [2011.11903]. The topological backbone of the theory is duality. In the cubical model, \[ \operatorname{rank}\phi_*+\operatorname{rank}\psi_*=D, \qquad D=\binom di, \] and in the permutohedral model, \[ \mathcal B_k(Q)+\mathcal B_{d-k}(Q^c)=\operatorname{rank}H_k(\mathbb T^d). \] These identities are the higher-dimensional analogue of planar primal-dual duality in ordinary percolation [2011.11903]. Sharp-threshold theorems show that, for fixed median threshold functions \(\lambda_N\), there is a transition from nonexistence of giant cycles to surjectivity onto ambient homology. For plaquettes, \[ \begin{cases} \mathbb P_{\lambda_N^{\square}-\epsilon}(A^{\square})\to 0,\\[2mm] \mathbb P_{\lambda_N^{\square}+\epsilon}(S^{\square})\to 1, \end{cases} \qquad N\to\infty, \] and an analogous statement holds for permutohedra [2011.11903]. The middle-dimensional self-dual case yields exact thresholds. If \(d=2i\), then for plaquettes, \[ p_c^{\square}(i,2i)=\frac12, \] and for permutohedra, \[ \lim_{N\to\infty}\lambda_{d/2}^{\hexagon}(N,d)=\frac12. \] These are finite-volume higher-dimensional analogues of Kesten’s \(p_c=\tfrac12\) theorems for bond percolation on the square lattice and site percolation on the triangular lattice [2011.11903]. At the extremal homological dimensions, exact constants are again available. For plaquettes, \[ p_c^{\square}(1,d)=\hat p_c(d), \qquad p_c^{\square}(d-1,d)=1-\hat p_c(d), \] while for permutohedra, \[ \lim_{N\to\infty}\lambda_1^{\hexagon}(N,d)=p_c(\mathcal A_d^*), \qquad \lim_{N\to\infty}\lambda_{d-1}^{\hexagon}(N,d)=1-p_c(\mathcal A_d^*). \] This embeds homological percolation within the classical threshold landscape while preserving genuinely higher-dimensional content [2011.11903]. ## 4. Euler characteristic, persistent homology, and empirical threshold proxies A major empirical development is the observed relation between homological-percolation thresholds and zeros of the expected Euler characteristic curve. For a filtration \(\{X_t\}\), the Euler characteristic is \[ \chi(X) := \sum_k (-1)^k \beta_k(X), \] the Euler characteristic curve is \[ \chi(t) := \chi(X_t), \] and the expected Euler characteristic curve is \[ \bar\chi(t):=\mean{\chi(t)}. \] In four different models on \(T^d\)—site percolation on cubical and permutahedral lattices, the Poisson–Boolean model, and Gaussian random fields—the simulations strongly indicate that the zeros of \(\bar\chi\) approximate the critical values for homological percolation [1910.10146]. The common empirical observation is that, in all tested models and in dimensions \(d=2,3,4\), \(\bar\chi(t)\) has exactly \((d-1)\) zeros, \[ t_1^{\mathrm{ec},\ldots,t_{d-1}^{\mathrm{ec}, \] and the giant-cycle thresholds satisfy \[ t_k^{\mathrm{perc} \approx t_k^{\mathrm{ec}. \] Because the limiting thresholds are not directly accessible in simulation, the paper uses the birth time of the first giant \(k\)-cycle as a finite-size proxy for \(t_k^{\mathrm{perc}}\) [1910.10146]. The discrete cubical model has \[ \bar\chi_Q(p) := \mean{\chi(Q(n,p))}= n \sum_{k=0}^d (-1)^k\binom{d}{k} (1-(1-p)^{2^{d-k}), \] while the permutahedral model has \[ \bar\chi_P(p) := \mean{\chi(P(n,p))} = n\sum\limits_{k=0}^d (-1)^{d-k}\left(1- (1-p)^{k+1}\right) \sum\limits_{j=0}^{k+1}(-1)^{k+1-j} \binom{k}{j}j^{d+1}. \] For the Poisson–Boolean model, \[ \bar\chi_B(\lambda) := \mean{\chi(B(n,\lambda))} = n e^{-\lambda}\left(1+\sum_{k=1}^{d-1}A_{d,k} \lambda^k\right), \] and for Gaussian random fields, \[ \bar\chi_G(\alpha) := \mean{\chi(G(\alpha))} = \frac{2}{\omega_d} (2\pi)^{-\frac{d+1}{2} H_{d-1}(-\alpha) e^{-\alpha^2/2}. \] These formulas furnish analytically or numerically accessible threshold proxies in settings where direct critical-value analysis is difficult [1910.10146]. The discrepancy \[ \Delta_k := (t^{\text{perc}_k - t^{\text{ec}_k) \] exhibits a sign pattern in dimensions \(d\ge 3\): \[ t_k^{\mathrm{perc} \le t_k^{\mathrm{ec} \quad \text{for } kd/2. \] In symmetric even-dimensional models, the middle-dimensional threshold satisfies \[ \Delta_{d/2}=0 \] by symmetry and duality in the permutahedral case, and up to tiny numerical or discretization error in the zero-mean Gaussian case [1910.10146]. Persistent homology supplies the operational language for distinguishing giant from small cycles. Giant cycles are persistent-homology features that are ambiently essential in the torus, while small cycles are born and die within the filtration but are trivial in the ambient space. In the simulations, giant classes are detected by checking which persistent-homology classes map to nontrivial classes in \(H_k(T^d)\), and the first birth among them is taken as the finite-size percolation time [1910.10146]. A related empirical observation appears in two-dimensional continuum percolation with disks. There, the union \[ \mathcal D(r)=\bigcup_{i=1}^N D_i(r) \] is analyzed via the alpha complex \(\mathcal C(r)\), with filtration parameter \[ \eta(N,r)=N\pi r^2. \] The longest-persisting \(H_1\) invariant is observed to be born at or near the percolation transition \[ \eta_c \approx 1.128, \] with the mean birth value moving closer to \(\eta_c\) as \(N\) increases [1804.07733]. This does not define giant torus-wrapping cycles, but it reinforces the broader idea that persistent topological signal localizes near percolative criticality. ## 5. Variants beyond the toroidal giant-cycle paradigm Not all topological percolation models use the toroidal ambient-essential definition, but several closely related theories broaden the scope of homological percolation. One codimension-one theory studies a random cubical set in \(\mathbb R^d\) formed from open \((d-1)\)-faces. In this model, holes are bounded connected components of the complement, equivalently generators of \(H_{d-1}\) in finite windows, and a graph is constructed whose vertices are holes and whose edges encode shared boundary faces. The resulting hole percolation threshold satisfies \[ p_c^{\rm face}(d)\le p_c^{\rm hole}(d)\le 1-p_c^{\rm bond}(d), \] implying \[ 00\), then the number \(N_\infty\) of infinite hole clusters obeys \[ N_\infty=1 \quad \text{almost surely,} \] and for \(p>p_c^{\rm hole}\), \[ P_p\bigl(x^* \overset{\text{hole}{\longleftrightarrow} y^*\bigr)\ge c \] for any dual-lattice points \(x^*,y^*\) [1809.07490]. This is homological in a strict sense because the percolating objects are codimension-one homology generators, though it is not formulated through inclusion into a torus. A different finite-graph variant defines homological triviality via coverings. For Bernoulli edge percolation on a sequence \((\mathcal G_t)\) covered by an infinite graph \(\mathcal H\), the lower cycle erasure threshold is \[ p_E^0=\sup\left\{p\in[0,1]:\lim_{t\to\infty}{\bf P}_E(t,p)=0\right\}, \] where \[ {\bf P}_E(t,p)=P\bigl(H_1(f_t,p)\neq 0\bigr). \] The principal theorem states that \[ \frac{d_t}{\ln n_t}\to\infty \quad\Longrightarrow\quad p_E^0=p_{\rm c}(\mathcal H), \] while \[ \frac{d_t}{\ln n_t}\to 0 \quad\Longrightarrow\quad p_E^0=0. \] Here \(d_t\) is the homological distance, the size of the shortest nontrivial cycle on \(\mathcal G_t\) [2011.02603]. This formulation identifies the growth rate of the shortest cover-detectable nontrivial cycle as the key parameter governing homological change. Hyperbolic percolation yields another homological threshold proxy. For finite local quotients \(G_t(m)\) of hyperbolic tilings, the induced first homology of the random open subgraph is \[ H_1(G_\varepsilon)=\ker \partial_1^\varepsilon / \operatorname{Im}\partial_2^\varepsilon, \] and the rank difference function \[ D(p)=\limsup_{t\to\infty}\mathbb E_p\!\left(\frac{\operatorname{rank} G_{t,\bar\varepsilon}-\operatorname{rank} G_{t,\varepsilon}}{|E_t|}\right) \] leads to the homological upper bound \[ p_c(G(m))\le p_h, \qquad p_h = \sup\left\{ p\in[0,1/2] : D(p)+p-\frac{2}{m}=0 \right\}. \] For \(G(5)\), the explicit bound obtained is \[ p_c(G(5))\le 0.299973. \] This approach does not define giant torus classes, but it uses the emergence of nontrivial first homology in finite approximants as a finite-volume signature of supercriticality [1408.4031]. A more recent continuum framework for random simplicial complexes proves a sharp phase transition for up-connectivity of \(q\)-simplices through \((q+1)\)-simplices, with critical intensities \[ 0<\beta_c^{(0)} \le \beta_c^{(1)} \le \cdots \le \beta_c^{(\alpha-1)} <\infty, \] and exponential decay below threshold together with a supercritical linear lower bound [2506.15038]. This is not homological percolation in the strict sense, since it concerns higher-order connectivity rather than homology groups, but it provides a rigorous higher-dimensional percolation framework adjacent to the homological theory. ## 6. Simplicial complexes, growth phenomena, and open problems In growing simplicial complexes, homological percolation transitions have been studied through Betti numbers rather than ambient torus classes. A minimal model with growth and preferential attachment exhibits two successive transitions controlled by the first and second Betti numbers. The giant cluster size scales as \[ G(p,t)\approx G(p)\, t, \] the first Betti number as \[ B_1(t)\approx b_1(p)\, t, \] and the second Betti number as \[ B_2(t)\approx b_2(p)\, t^{0.70}. \] For \(a=0.1\), the first two critical values are reported as \[ p_{c0}\approx 0.0031, \qquad p_{c1}\approx 0.0031, \qquad p_{c2}\approx 0.053(3), \] with essential singularities \[ G(p)\sim \exp[-\alpha_0(p-p_{c0})^{-\beta_0}], \] \[ b_1(p)\sim \exp[-\alpha_1(p-p_{c1})^{-\beta_1}], \] \[ b_2(p)\sim \exp[-\alpha_2(p-p_{c2})^{-\beta_2}], \] and \[ \beta_0 \approx 0.44,\qquad \beta_1 \approx 0.44,\qquad \beta_2 \approx 0.99. \] These transitions are described as infinite order of the Berezinskii–Kosterlitz–Thouless type [2010.12224]. This line of work differs from the toroidal giant-cycle framework because the order parameters are the Betti numbers of the giant cluster rather than inclusion-induced images in ambient homology. It also exhibits delocalization: \(B_1\) continues to increase even after \(B_2\) appears, in contrast to the Kahle localization observed in static random simplicial complexes. The proposed mechanism is that delocalization arises when the merging rate of two-dimensional simplexes is less than the birth rate of isolated simplexes [2010.12224]. This suggests that growth fundamentally alters topological criticality. Across the subject, several open problems recur. In the continuum toroidal theory, the conjectured sharp picture is \[ \lambda_{0,k}=\lambda_{1,k}=: \lambda_k, \qquad \lambda_c = \lambda_1 < \lambda_2 < \cdots < \lambda_{d-1} = \bar \lambda_c, \] but only the case \(k=1\) is proved exactly [2005.14011]. In the Euler-characteristic program, the central relation \[ t_k^{\mathrm{perc} \approx t_k^{\mathrm{ec} \] is supported experimentally rather than theoretically, and the limiting behavior of \[ \Delta_k \] remains unresolved except in symmetry-forced cases [1910.10146]. In finite toroidal lattice models, it remains open whether the threshold functions converge to true constants in all dimensions and homological degrees [2011.11903]. In covering-space graph models, the logarithmic regime suggests \(p_E^0