{
  "id": "enwiki-001370071433",
  "slug": "topological-recursion",
  "title": "Topological recursion",
  "qid": "",
  "lead": "In mathematics, topological recursion is a recursive definition of invariants of spectral curves. It has applications in enumerative geometry, random matrix theory, mathematical physics, string theory, knot theory.",
  "body": "In mathematics, topological recursion is a recursive definition of invariants of spectral curves. It has applications in enumerative geometry, random matrix theory, mathematical physics, string theory, knot theory. \\begin{align}\\omega_{g,n}(z_1,z_2,\\dots,z_{n}) &=\\sum_{a=\\text{branchpoints}} \\operatorname{Res}_{z\\to a} K(z_1,z,\\sigma_a(z)) \\Big( \\omega_{g-1,n+1}(z,\\sigma_a(z),z_2,\\dots,z_n) \\\\ &\\qquad\\qquad\\qquad + \\mathop{{\\sum}'}_{\\overset{g_1+g_2=g}{I_1\\uplus I_2=\\{z_2,\\dots,z_n\\} }} \\omega_{g_1,1+\\# I_1}(z,I_1)\\omega_{g_2,1+\\# I_2}(\\sigma_a(z),I_2) \\Big) \\end{align} where K(z_1,z_2,z_3) is called the recursion kernel: K(z_1,z_2,z_3) = \\frac{\\frac12 \\int_{z'=z_3}^{z_2} \\omega_{0,2}(z_1,z')}{\\omega_{0,1}(z_2)-\\omega_{0,1}(z_3)} and \\sigma_a is the local Galois involution near a branch point a , it is such that x(\\sigma_a(z))=x(z) . The primed sum {\\sum}' means excluding the two terms (g_1,I_1)=(0,\\emptyset) and (g_2,I_2)=(0,\\emptyset) . For n=0 and 2g-2>0 : F_g = \\omega_{g,0} = \\frac{1}{2-2g}\\ \\sum_{a=\\text{branchpoints}} \\operatorname{Res}_{z\\to a} F_{0,1}(z) \\omega_{g,1}(z) with dF_{0,1}=\\omega_{0,1} any antiderivative of \\omega_{0,1} . The definition of F_0=\\omega_{0,0} and F_1=\\omega_{1,0} is more involved and can be found in the original article of Eynard-Orantin. \\sum_{a=\\text{branchpoints}} \\operatorname{Res}_{z\\to a} F_{0,1}(z)\\ \\omega_{g,n+1}(z_1,\\dots,z_n,z) = (2g-2+n) \\omega_{g,n}(z_1,\\dots,z_n) where dF_{0,1}=\\omega_{0,1} . Loop equations: The following forms have no poles at branchpoints \\sum_{z\\in x^{-1}(x)} \\omega_{g,n+1}(z,z_1,\\dots,z_n) \\sum_{\\{z\\neq z'\\} \\subset x^{-1}(x)} \\Big(\\omega_{g,n+1}(z,z',z_2,\\dots,z_n) + \\sum_{\\overset{g_1+g_2=g}{I_1\\uplus I_2=\\{z_2,\\dots,z_n\\} }} \\omega_{g_1,1+\\# I_1}(z,I_1)\\omega_{g_2,1+\\# I_2}(z',I_2) \\Big) where the sum has no prime, i.e. no term excluded. Deformations: The \\omega_{g,n} satisfy deformation equations Limits: given a family of spectral curves \\mathcal S_t , whose limit as t\\to 0 is a singular curve, resolved by rescaling by a power of t^\\mu , then \\lim_{t\\to 0} t^{(2-2g-n)\\mu}\\omega_{g,n}(\\mathcal S_t) = \\omega_{g,n}(\\lim_{t\\to 0} t^\\mu \\mathcal S_t) . Symplectic invariance: In the case where \\Sigma is a compact algebraic curve with a marking of a symplectic basis of cycles, x is meromorphic and \\omega_{0,1}=ydx is meromorphic and \\omega_{0,2}=B is the fundamental second kind differential normalized on the marking, then the spectral curve \\mathcal S=(\\Sigma,\\mathbb C,x,ydx,B) and \\tilde{\\mathcal S}=(\\Sigma,\\mathbb C,y,-xdy,B) , have the same F_g shifted by some terms. Modular properties: In the case where \\Sigma is a compact algebraic curve with a marking of a symplectic basis of cycles, and \\omega_{0,2}=B is the fundamental second kind differential normalized on the marking, then the invariants \\omega_{g,n} are quasi-modular forms under the modular group of marking changes. The invariants \\omega_{g,n} satisfy BCOV equations. In case the branchpoints are not simple, the definition is amended as follows (simple branchpoints correspond to k=2): \\begin{align} \\omega_{g,n}(z_1,z_2,\\dots,z_{n}) = &\\sum_{a=\\text{branchpoints}} \\operatorname{Res}_{z\\to a} \\sum_{k=2}^{{\\rm order}_x(a)} \\sum_{J \\subset x^{-1}(x(z))\\setminus\\{z\\},\\,\\# J=k-1} K_k(z_1,z,J) \\\\ & \\qquad \\cdot\\sum_{J_1,\\dots , J_\\ell \\vdash J\\cup\\{z\\}} \\sum'_{\\overset{g_1+\\dots+g_\\ell=g+\\ell-k}{I_1\\uplus \\dots I_\\ell=\\{z_2,\\dots,z_n\\} }} \\prod_{i=1}^l \\omega_{g_i,\\# J_i+\\# I_i}(J_i,I_i) \\end{align} The first sum is over partitions J_1,\\dots,J_\\ell of J\\cup\\{z\\} with non empty parts J_i\\neq \\emptyset , and in the second sum, the prime means excluding all terms such that (g_i,\\# J_i+\\# I_i)=(0,1) . K_k is called the recursion kernel: K_k(z_0,z_1,\\dots,z_k) = \\frac{\\int_{z'=*}^{z_1} \\omega_{0,2}(z_0,z')}{\\prod_{i=2}^k (\\omega_{0,1}(z_1)-\\omega_{0,1}(z_i))} The base point * of the integral in the numerator can be chosen arbitrarily in a vicinity of the branchpoint, the invariants \\omega_{g,n} will not depend on it. = 2^{3g-3+n} &\\sum_{G=\\text{Graphs}} \\frac{1}{\\#\\text{Aut}(G)} \\int_{\\left(\\prod_{v=\\text{vertices}} {\\overline{\\mathcal M}}_{g_v,n_v} \\right)}\\,\\, \\prod_{v=\\text{vertices}} e^{\\sum_k \\hat t_{\\sigma(v),k} \\kappa_k} \\\\ &\\prod_{(p,p')=\\text{nodal points}} \\left(\\sum_{d,d'} B_{\\sigma(p),2d;\\sigma(p'),2d'} \\psi_p^d \\psi_{p'}^{d'}\\right) \\prod_{p_i=\\text{marked points}\\, i=1,\\dots,n} \\left(\\sum_{d_i} \\psi_{p_i}^{d_i} d\\xi_{\\sigma(p_i),d_i}(z_i) \\right) \\end{align} where the sum is over dual graphs of stable nodal Riemann surfaces of total arithmetic genus g , and n smooth labeled marked points p_1,\\dots,p_n , and equipped with a map \\sigma:\\{\\text{vertices}\\}\\to \\{\\text{branchpoints}\\} . \\psi_p=c_1(\\mathcal L_p) is the Chern class of the cotangent line bundle \\mathcal L_p whose fiber is the cotangent plane at p . \\kappa_k is the k th Mumford's kappa class. The coefficients \\hat t_{a,k} , B_{a,k;a',k'} , d\\xi_{a,k}(z) , are the Taylor expansion coefficients of \\omega_{0,1} and \\omega_{0,2} in the vicinity of branchpoints as follows: in the vicinity of a branchpoint a (assumed simple), a local coordinate is \\zeta_a(z)=\\sqrt{x(z)-a} . The Taylor expansion of \\omega_{0,2}(z,z') near branchpoints z\\to a , z'\\to a' defines the coefficients B_{a,d;a',d'} \\omega_{0,2}(z,z') \\mathop _{z\\to a,\\ z'\\to a'} \\left( \\frac{\\delta_{a,a'} }{(\\zeta_a(z)-\\zeta_{a'}(z'))^2}+ 2\\pi \\sum_{d,d'=0}^\\infty \\frac{B_{a,d;a',d'}}{\\Gamma(\\frac{d+1}{2})\\Gamma(\\frac{d'+1}{2})}\\, \\zeta_a(z)^d \\zeta_{a'}(z')^{d'} \\right) d\\zeta_a(z)d\\zeta_{a'}(z') . The Taylor expansion at z'\\to a , defines the 1-forms coefficients d\\xi_{a,d}(z) d\\xi_{a,d}(z) = \\frac{-\\Gamma(d+\\frac12)}{\\Gamma(\\frac12)} \\operatorname{Res}_{z'\\to a} (x(z')-a)^{-d-\\frac12}\\omega_{0,2}(z,z') whose Taylor expansion near a branchpoint a' is d\\xi_{a,d}(z) \\mathop _{z\\to a'} \\frac{-\\delta_{a,a'} (2d+1)!! d\\zeta_a(z)}{2^d \\zeta_a(z)^{2d+2}}+ \\sum_{k=0}^\\infty \\frac{B_{a,2d;a',2k} 2^{k+1}}{(2k-1)!!}\\zeta_{a'}(z)^{2k} d\\zeta_{a'}(z) . Write also the Taylor expansion of \\omega_{0,1} \\omega_{0,1}(z) \\mathop _{z\\to a} \\sum_{k=0}^\\infty t_{a,k}\\ \\frac{\\Gamma(\\frac12)}{(k+1)\\Gamma(\\frac{k+1}{2})}\\ \\zeta_{a}(z)^{k} d\\zeta_{a}(z) . Equivalently, the coefficients t_{a,k} can be found from expansion coefficients of the Laplace transform, and the coefficients \\hat t_{a,k} are the expansion coefficients of the log of the Laplace transform \\int_{x(z)-x(a)\\in \\mathbb R_+} \\omega_{0,1}(z) e^{-u x(z)} = \\frac{e^{-ux(a)}\\sqrt\\pi}{2 u^{3/2}} \\sum_{k=0}^\\infty t_{a,k} u^{-k} = \\frac{e^{-ux(a)}\\sqrt\\pi}{2 u^{3/2}} e^{-\\sum_{k=0}^\\infty \\hat t_{a,k} u^{-k}} . For example, we have \\omega_{0,3}(z_1,z_2,z_3) = \\sum_a e^{\\hat t_{a,0}} d\\xi_{a,0}(z_1)d\\xi_{a,0}(z_2)d\\xi_{a,0}(z_3). \\omega_{1,1}(z) = 2 \\sum_a e^{\\hat t_{a,0}} \\left( \\frac{1}{24} d\\xi_{a,1}(z) + \\frac{\\hat t_{a,1}}{24} d\\xi_{a,0}(z) +\\frac12 B_{a,0;a,0} d\\xi_{a,0}(z)\\right). The formula (*) generalizes ELSV formula as well as Mumford's formula and Mariño-Vafa formula. \\left(\\mathbb C; \\ \\mathbb C;\\ x: z\\mapsto z^2 ;\\ \\omega_{0,1}(z)=\\frac{4}{\\pi} z \\sin{(\\pi z)} dz;\\,\\omega_{0,2}(z_1,z_2) = \\frac{dz_1 dz_2}{(z_1-z_2)^2}\\right) the n-form \\omega_{g,n} = d_1 \\dots d_n F_{g,n} is the Laplace transform of the Weil-Petersson volume F_{g,n}(z_1,\\dots,z_n) = \\int_0^\\infty e^{-z_1L_1} dL_1 \\dots \\int_0^\\infty e^{-z_nL_n} dL_n \\quad \\int_{\\mathcal M_{g,n}(L_1,\\dots,L_n)} w where \\mathcal M_{g,n}(L_1,\\dots,L_n) is the moduli space of hyperbolic surfaces of genus g with n geodesic boundaries of respective lengths L_1,\\dots,L_n , and w is the Weil-Petersson volume form. The topological recursion for the n-forms \\omega_{g,n}(z_1,\\dots,z_n) , is then equivalent to Mirzakhani's recursion. \\left(\\mathbb C; \\ \\mathbb C;\\ x: z\\mapsto z^2 ;\\ \\omega_{0,1}(z)=2z^2 dz;\\,\\omega_{0,2}(z_1,z_2) = \\frac{dz_1 dz_2}{(z_1-z_2)^2}\\right) the n-form \\omega_{g,n}= d_1 \\dots d_n F_{g,n} is F_{g,n}(z_1,\\dots,z_n) = 2^{2-2g-n}\\sum_{d_1+\\dots+d_n=3g-3+n} \\prod_{i=1}^n \\frac{(2d_i-1)!! }{z_i^{2d_i+1}} \\quad \\left\\langle\\tau_{d_1}\\dots\\tau_{d_n}\\right\\rangle_g where \\left\\langle\\tau_{d_1}\\dots\\tau_{d_n}\\right\\rangle_g is the Witten-Kontsevich intersection number of Chern classes of cotangent line bundles in the compactified moduli space of Riemann surfaces of genus g with n smooth marked points. \\left(\\mathbb C; \\ \\mathbb C;\\ x:z \\mapsto -z+\\ln{z} ;\\ \\omega_{0,1}(z)=(1-z) dz;\\,\\omega_{0,2}(z_1,z_2) = \\frac{dz_1 dz_2}{(z_1-z_2)^2}\\right) the n-form \\omega_{g,n}= d_1 \\dots d_n F_{g,n} is F_{g,n}(z_1,\\dots,z_n) = \\sum_{\\ell(\\mu)\\leq n} m_\\mu(e^{x(z_1)},\\dots,e^{x(z_n)}) \\quad h_{g,\\mu_1,\\dots,\\mu_n} where h_{g,\\mu} is the connected simple Hurwitz number of genus g with ramification \\mu=(\\mu_1,\\dots,\\mu_n) : the number of branch covers of the Riemann sphere by a genus g connected surface, with 2g-2+n simple ramification points, and one point with ramification profile given by the partition \\mu . Its mirror manifold is singular over a complex plane curve \\Sigma given by a polynomial equation P(e^x,e^y)=0 , whose coefficients are functions of the Kähler moduli. For the choice of spectral curve \\left(\\Sigma; \\ \\mathbb C^*;\\ x ;\\ \\omega_{0,1}=y dx;\\,\\omega_{0,2} \\right) with \\omega_{0,2} the fundamental second kind differential on \\Sigma , According to the BKMP conjecture, the n-form \\omega_{g,n}= d_1 \\dots d_n F_{g,n} is F_{g,n}(z_1,\\dots,z_n) = \\sum_{\\mathbf d\\in H_2(\\mathfrak X,\\mathbb Z)} \\sum_{\\mu_1,\\dots,\\mu_n\\in H_1(\\mathcal L,\\mathbb Z)} t^d \\prod_{i=1}^n e^{x(z_i)} \\mathcal N_{g}(\\mathfrak X,\\mathcal L;\\mathbf d,\\mu_1,\\dots,\\mu_n) where \\mathcal N_{g}(\\mathfrak X,\\mathcal L;\\mathbf d,\\mu_1,\\dots,\\mu_n) = \\int_{[{\\overline{\\mathcal M}}_{g,n}(\\mathfrak X,\\mathcal L, \\mathbf d,\\mu_1,\\dots,\\mu_n)]^{\\rm vir}} 1 is the genus g Gromov–Witten number, representing the number of holomorphic maps of a surface of genus g into \\mathfrak X , with n boundaries mapped to a special Lagrangian submanifold \\mathcal L . \\mathbf d=(d_1,\\dots,d_{b_2(\\mathfrak X)}) is the 2nd relative homology class of the surface's image, and \\mu_i\\in H_1(\\mathcal L,\\mathbb Z) are homology classes (winding number) of the boundary images. The BKMP conjecture has since then been proven.",
  "categories": [
    "20383304",
    "20476451",
    "20476469",
    "22945460",
    "28607422"
  ],
  "license": "CC BY-SA 4.0",
  "source_uri": "https://en.wikipedia.org/w/index.php?oldid=1370071433",
  "content_hash": "",
  "created_at": null,
  "updated_at": 1787090156.0,
  "claims": [],
  "contradictions": [],
  "confidence": 0.6,
  "csl": {
    "id": "enwiki-001370071433",
    "type": "webpage",
    "title": "Topological recursion",
    "author": [
      {
        "family": "Anticloud FZ LLE",
        "literal": "Anticloud FZ LLE"
      }
    ],
    "container-title": "encyclopegia.org",
    "publisher": "Anticloud FZ LLE",
    "publisher-place": "Dubai, United Arab Emirates",
    "URL": "https://encyclopegia.org/a/topological-recursion",
    "issued": {
      "date-parts": [
        [
          2026,
          11,
          20
        ]
      ]
    },
    "accessed": {
      "date-parts": [
        [
          2026,
          11,
          20
        ]
      ]
    },
    "license": "CC BY-SA 4.0",
    "DOI": "eo1:enwiki-001370071433",
    "isBasedOn": "https://en.wikipedia.org/w/index.php?oldid=1370071433",
    "note": "Immutable snapshot enwiki-001370071433"
  }
}