{
  "id": "persistent-homology",
  "title": "Persistent Homology",
  "lang": "en",
  "source_lang": "en",
  "summary": "Persistent homology tracks topological features across a nested family of spaces, describing their appearance and disappearance through intervals called persistence barcodes.",
  "category": "mathematics",
  "subcategory": "geometry",
  "type": "concept",
  "aliases": [
    "Persistence homology"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/persistent-homology/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/persistent-homology/",
      "json": "https://aiwiki.page/zh/persistent-homology.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Persistent homology is a method in algebraic topology that studies how connected components, loops, cavities, and higher-dimensional features appear and disappear as a space changes through a nested sequence. Rather than describing a data set at a single resolution, it follows its homology across many resolutions. The resulting birth and death values are commonly represented by a barcode or persistence diagram. Persistent homology is a central tool of topological data analysis, where it provides multiscale descriptions of the shape of data."
    },
    {
      "heading": "From homology to persistence",
      "text": "Homology assigns algebraic objects to a space that describe particular kinds of topological structure. With coefficients in a field \\(\\mathbb F\\), its homology groups are vector spaces:\n\n- \\(H_0\\) describes connected components.\n- \\(H_1\\) describes independent loops.\n- \\(H_2\\) describes enclosed two-dimensional shells, such as the surface surrounding a three-dimensional cavity.\n- Higher homology groups describe analogous higher-dimensional features.\n\nThe dimension of \\(H_k\\) is the \\(k\\)-th Betti number. Ordinary homology describes a particular space, but data often do not determine one uniquely. A point cloud, for example, has no connections between its points until a rule specifies which nearby points should be connected.\n\nPersistent homology addresses this scale dependence by considering a filtration: a nested family of spaces or combinatorial complexes,\n\n\\[K_s\\subseteq K_t \\qquad\\text{whenever }s\\leq t.\\]\n\nThe parameter may represent distance, a function threshold, image intensity, or another quantity. This topological meaning of filtration is distinct from a filtration of information in probability theory.\n\nEach inclusion induces a linear map\n\n\\[\\varphi_k^{s,t}:H_k(K_s;\\mathbb F)\\longrightarrow H_k(K_t;\\mathbb F).\\]\n\nThese maps record which classes remain detectable as the parameter increases. A class can disappear because previously separate components merge, because a loop becomes filled by faces, or because a higher-dimensional cycle becomes a boundary.\n\nThe important information is therefore not just the homology at each parameter value, but also the maps connecting those homology spaces."
    },
    {
      "heading": "Building filtrations",
      "text": "Point-cloud filtrations\n\nFor a finite set of points in a metric space, a filtration can be constructed by gradually enlarging neighborhoods around the points.\n\nThe Čech complex at radius \\(r\\) contains a simplex whenever the corresponding radius-\\(r\\) balls have a common intersection. Under the usual hypotheses of the nerve theorem—for example, a finite collection of open convex balls in Euclidean space—this complex has the same homotopy type as the union of the balls.\n\nThe Vietoris–Rips complex at threshold \\(r\\) contains a simplex whenever every pair of its vertices has distance at most \\(r\\). It depends only on pairwise distances and is therefore convenient for general metric data. A triangle enters when all three edges are present, even if the corresponding balls have no common intersection.\n\nČech and Vietoris–Rips filtrations use different geometric criteria. Their scale parameters must be compared carefully, since conventions also differ over whether the parameter denotes a ball radius or an edge-distance threshold.\n\nFor Euclidean point clouds, alpha complexes use Delaunay geometry to obtain smaller complexes related to unions of balls. Their practical advantage can diminish in high ambient dimension.\n\nFunction and image filtrations\n\nA real-valued continuous function \\(f:X\\to\\mathbb R\\) gives a sublevel-set filtration\n\n\\[X_t=f^{-1}((-\\infty,t]).\\]\n\nAs the threshold rises, more of the space enters the filtration. Superlevel-set filtrations instead follow regions above a threshold, with the indexing direction adjusted appropriately.\n\nThese constructions apply to height functions, density estimates, energy landscapes, and scalar fields. For digital images and volumetric data, cubical complexes often provide a natural representation: pixels, voxels, and their faces enter according to a specified threshold rule.\n\nThe choice of filtration is part of the model. Different filtrations of the same data can reveal different structures and need not give equivalent results."
    },
    {
      "heading": "Persistence modules and interval decomposition",
      "text": "For a fixed homological degree \\(k\\), the spaces \\(H_k(K_t;\\mathbb F)\\), together with their induced maps, form a persistence module. The maps satisfy\n\n\\[\\varphi_k^{t,t}=\\operatorname{id}, \\qquad \\varphi_k^{s,u} = \\varphi_k^{t,u}\\circ\\varphi_k^{s,t} \\quad\\text{for }s\\leq t\\leq u.\\]\n\nFor \\(s\\leq t\\), the persistent homology group is often defined as\n\n\\[H_k^{s,t} = \\operatorname{im}\\varphi_k^{s,t}.\\]\n\nIts dimension counts the independent classes present at \\(s\\) that remain detectable at \\(t\\). These dimensions form the persistent rank invariant.\n\nFor a finite filtration over a field, the persistence module decomposes into a direct sum of interval modules. An interval module contributes one copy of \\(\\mathbb F\\) at parameter values within its interval, zero outside it, and identity maps between its nonzero spaces. Suitable interval-decomposition theorems also hold for pointwise finite-dimensional modules indexed by the real line.\n\nEach interval describes:\n\n- a birth, when a class first appears;\n- a death, when it ceases to persist;\n- a persistence, usually the difference between finite death and birth values.\n\nAn interval may continue indefinitely. In a finite filtration, such an interval means that the feature survives in the final complex, not necessarily that it would survive every possible extension of the filtration.\n\nThe interval multiset is invariant up to the relevant notion of module isomorphism. Individual cycle representatives, however, are generally not canonical.\n\nBarcodes and persistence diagrams\n\nA barcode displays each interval as a horizontal bar. A persistence diagram records a finite interval with birth \\(b\\) and death \\(d\\) as a point \\((b,d)\\), counted with multiplicity. Essential classes with infinite death are handled separately or represented using extended coordinates.\n\nPoints far from the diagonal \\(b=d\\) have long persistence. Points near it have short persistence. Zero-length intervals are normally omitted from the diagram.\n\nLong intervals are often useful indicators of structure that remains visible over a broad range of scales. They are not automatically meaningful signals, and short intervals are not automatically noise: interpretation depends on the filtration, sampling process, and scientific question."
    },
    {
      "heading": "A simple example",
      "text": "Consider three points at the vertices of an equilateral triangle of side length \\(a\\).\n\nIn the Vietoris–Rips filtration with edge threshold \\(r\\), there are initially three connected components. At \\(r=a\\), all three edges and the filled triangle enter simultaneously. Two components merge into the remaining component, but no positive-length one-dimensional interval appears: the triangular loop is filled at the same parameter value at which it forms.\n\nIn the Čech filtration with ball radius \\(r\\), each pair of balls first intersects at \\(r=a/2\\). The three edges then form a loop. The filled triangle enters only when all three balls intersect, at\n\n\\[r=\\frac{a}{\\sqrt{3}}.\\]\n\nUnder a closed-ball convention, the loop therefore contributes the interval\n\n\\[\\left[\\frac a2,\\frac a{\\sqrt{3}}\\right).\\]\n\nEndpoint conventions can change with the filtration definition, but the positive lifetime in this example does not.\n\nThis contrast shows why specifying the filtration matters as much as specifying the data."
    },
    {
      "heading": "Computation",
      "text": "For a finite filtered simplicial complex, a standard algorithm orders simplices by filtration value, placing each face before its cofaces when values are tied. It then forms a boundary matrix over the chosen coefficient field and reduces columns using additions compatible with the filtration order.\n\nThe reduced matrix identifies persistence pairings:\n\n- a simplex that creates a homology class begins an interval;\n- a later simplex paired with it ends that interval;\n- an unpaired creator corresponds to a class surviving to the end.\n\nThis resembles Gaussian elimination, but the ordering restrictions are essential for preserving filtration information. Coefficients in the two-element finite field are common because arithmetic is simple and orientation signs disappear. Different coefficient fields can nevertheless produce different homology and persistence.\n\nFor \\(N\\) simplices, standard matrix reduction has cubic worst-case arithmetic complexity and quadratic worst-case storage requirements. Practical performance often improves substantially through sparse representations, clearing, compression, cohomological methods, and implicit construction of complexes.\n\nThe number of simplices can itself be the dominant obstacle. A Vietoris–Rips complex on \\(n\\) points may contain up to \\(2^n-1\\) nonempty simplices. When computing \\(H_k\\), only simplices through dimension \\(k+1\\) are needed, but their number can still be large."
    },
    {
      "heading": "Stability",
      "text": "Stability theorems explain why persistent homology can tolerate small perturbations of suitable input data.\n\nThe bottleneck distance between persistence diagrams optimally matches their points using the maximum-coordinate distance. Points may also be matched to the diagonal, which represents the disappearance of short intervals. The distance from a point \\((b,d)\\) to the diagonal is \\((d-b)/2\\).\n\nFor suitably tame functions \\(f,g:X\\to\\mathbb R\\), their sublevel-set persistence diagrams satisfy\n\n\\[d_B\\bigl(\\operatorname{Dgm}_k(f),\\operatorname{Dgm}_k(g)\\bigr) \\leq \\|f-g\\|_\\infty.\\]\n\nThus a small uniform change in function values cannot produce an arbitrarily large change in the diagram. Related results compare filtrations through interleavings and control changes caused by perturbations of metric data.\n\nThese guarantees have limits. Stability is relative to a specified input metric and construction; it does not imply robustness to every kind of contamination. An isolated outlier, for example, may generate a long-lived connected component. Nor does stability establish statistical significance or recover the topology of an unknown underlying space without additional assumptions."
    },
    {
      "heading": "Applications and statistical interpretation",
      "text": "Persistent homology is used to describe geometric and organizational patterns in point clouds, images, networks, materials, molecular configurations, and dynamical data. In degree zero, it is closely related to single-linkage clustering: components merge as distance thresholds increase. Higher degrees describe loops and cavities that cluster membership alone does not capture.\n\nIn time-series analysis, delay-coordinate point clouds can expose loop-like structure associated with recurrent behavior. In materials analysis, filtrations can characterize pores and spatial organization. In machine learning, barcodes and diagrams can be converted into numerical features through persistence landscapes, persistence images, kernels, or other summaries.\n\nStatistical interpretation requires assumptions beyond the algebraic construction. Recovery guarantees typically depend on sampling density, geometric regularity, and the observation model. Bootstrap procedures and other inferential methods can quantify uncertainty under appropriate conditions, but a bar’s length is not itself a probability or a significance level."
    },
    {
      "heading": "History and extensions",
      "text": "Persistent homology developed from earlier work on the topology of changing shapes and functions. Size theory in the early 1990s provided an important precursor. The 2002 paper Topological Persistence and Simplification, by Herbert Edelsbrunner, David Letscher, and Afra Zomorodian, established a widely influential computational formulation. Subsequent work clarified the algebraic classification of persistence modules, developed stability theory, and expanded practical algorithms.\n\nSeveral extensions broaden the original framework:\n\n- Extended persistence combines ordinary and relative homology to pair features that would otherwise remain essential.\n- Zigzag persistence allows maps to point in either direction, accommodating sequences with both additions and deletions.\n- Multiparameter persistence varies several parameters simultaneously, such as distance and density threshold. Unlike ordinary one-parameter persistence over a field, it generally has no complete classification by a multiset of intervals.\n- Persistent cohomology provides a dual computational viewpoint and supports applications such as circular coordinates."
    },
    {
      "heading": "Limitations",
      "text": "Persistent homology is a summary rather than a complete description of shape. Different spaces and data sets can produce identical barcodes, and even the full persistence module does not generally determine the original filtered space.\n\nIts results depend on the metric, filtration, coefficient field, and selected homological degrees. A loop detected after a particular embedding or preprocessing step may reflect that construction rather than an intrinsic property of the observations. The numerical parameterization also matters: an order-preserving transformation can retain the sequence of topological changes while substantially altering interval lengths.\n\nFinally, locating a detected feature in the original data is a separate problem. A barcode identifies when a class persists, but it does not uniquely specify a geometric cycle. Finding useful representatives, distinguishing artifacts from relevant structure, and relating topological summaries to domain-specific quantities require additional methods."
    }
  ],
  "links": [
    "algebraic-topology",
    "topological-data-analysis",
    "field-mathematics",
    "vector-space",
    "linear-map",
    "metric-space",
    "euclidean-space",
    "continuous-function",
    "direct-sum",
    "matrix",
    "gaussian-elimination",
    "finite-field",
    "cluster-analysis",
    "time-series",
    "machine-learning",
    "bootstrap-sampling"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}