{
  "id": "functional-mathematics",
  "title": "Functional (Mathematics)",
  "lang": "en",
  "source_lang": "en",
  "summary": "A functional is a scalar-valued mapping, typically defined on a space of functions, vectors, or other mathematical objects.",
  "category": "mathematics",
  "subcategory": "analysis",
  "type": "concept",
  "aliases": [
    "Scalar-valued functional"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/functional-mathematics/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/functional-mathematics/",
      "json": "https://aiwiki.page/zh/functional-mathematics.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "A functional is a function whose output is a scalar, usually a real number or a complex number, and whose input is typically a function or an element of a vector space. For example, a rule assigning each continuous function its integral over a fixed interval is a functional. Functionals are central to functional analysis and the calculus of variations, where entire functions are treated as variables to be studied, differentiated, or optimized."
    },
    {
      "heading": "Definition and terminology",
      "text": "A scalar-valued functional on a set \\(X\\) is a mapping\n\n\\[F:X\\longrightarrow \\mathbb K,\\]\n\nwhere \\(\\mathbb K\\) is usually \\(\\mathbb R\\) or \\(\\mathbb C\\). In the most common settings, \\(X\\) is a space of functions equipped with vector-space operations and some notion of convergence.\n\nThe distinction between a function and a functional is primarily one of emphasis, not a difference in their set-theoretic nature. Every functional is a function. The term functional highlights that its argument may itself be a function, or that its scalar output is being contrasted with a vector-valued output.\n\nNotation such as \\(F[f]\\) or \\(J[u]\\) is often used to emphasize a function-valued argument, but ordinary notation such as \\(F(f)\\) is equally valid. Square brackets have no intrinsic mathematical significance.\n\nAn operator commonly maps functions or vectors to other functions or vectors, whereas a functional produces a scalar. Thus, differentiation \\(f\\mapsto f'\\) is ordinarily called an operator, while \\(f\\mapsto f'(0)\\), on a suitable domain, is a functional. Usage varies: in parts of functional analysis, “functional” may implicitly mean a linear or continuous linear functional, so the surrounding definitions matter."
    },
    {
      "heading": "Basic examples",
      "text": "Evaluation\n\nFor a fixed point \\(t_0\\), the evaluation functional is\n\n\\[E_{t_0}(f)=f(t_0).\\]\n\nOn \\(C([a,b])\\), the space of continuous scalar-valued functions on a compact interval, this is a linear functional. With the uniform norm\n\n\\[\\|f\\|_\\infty=\\max_{t\\in[a,b]}|f(t)|,\\]\n\nit satisfies\n\n\\[|E_{t_0}(f)|\\leq \\|f\\|_\\infty.\\]\n\nIts meaning depends on the chosen function space. In the usual L^p spaces, elements are equivalence classes of functions that agree almost everywhere. Changing a function at a single point does not change its \\(L^p\\) element, so point evaluation generally does not define a functional on such a space.\n\nIntegration\n\nThe integral defines a linear functional:\n\n\\[I(f)=\\int_a^b f(t)\\,dt.\\]\n\nMore generally, for a fixed weight \\(w\\),\n\n\\[I_w(f)=\\int_a^b w(t)f(t)\\,dt\\]\n\nis linear on any vector space of functions for which the integral exists.\n\nIntegration illustrates why the domain must be specified. The same displayed formula can define a continuous functional on one function space and fail to be defined on all elements of another.\n\nNorms and energies\n\nA norm is a real-valued functional:\n\n\\[N(f)=\\|f\\|.\\]\n\nIt is generally nonlinear. In particular, absolute homogeneity,\n\n\\[N(\\alpha f)=|\\alpha|N(f),\\]\n\ndiffers from the homogeneity required for a linear functional.\n\nAnother common example is\n\n\\[Q(f)=\\int_a^b |f(t)|^2\\,dt.\\]\n\nThis is a nonnegative quadratic functional on \\(L^2([a,b])\\). Such expressions often measure energy, size, or error.\n\nExpectation\n\nOn a fixed probability space, expectation is a linear functional on the space of integrable random variables:\n\n\\[F(X)=\\mathbb E[X].\\]\n\nOn \\(L^1\\), it is continuous because\n\n\\[|\\mathbb E[X]|\\leq \\mathbb E[|X|].\\]\n\nThe requirement that the underlying probability measure be fixed is important: expectation depends on both the random variable and the measure used to integrate it."
    },
    {
      "heading": "Linear and continuous functionals",
      "text": "A linear functional on a vector space \\(X\\) over \\(\\mathbb K\\) satisfies\n\n\\[\\ell(\\alpha x+\\beta y) =\\alpha\\ell(x)+\\beta\\ell(y)\\]\n\nfor all \\(x,y\\in X\\) and \\(\\alpha,\\beta\\in\\mathbb K\\). It is therefore a linear map from \\(X\\) to its scalar field.\n\nIf \\(X\\) is a normed vector space, a linear functional is continuous if and only if it is bounded: there is a constant \\(C\\geq 0\\) such that\n\n\\[|\\ell(x)|\\leq C\\|x\\| \\qquad\\text{for every }x\\in X.\\]\n\nIts operator norm is\n\n\\[\\|\\ell\\|=\\sup_{\\|x\\|\\leq 1}|\\ell(x)|.\\]\n\nFor linear functionals, continuity at zero already implies continuity everywhere. This equivalence between continuity and boundedness is specific to the linear setting; a nonlinear functional may be continuous without being globally bounded.\n\nEvery linear functional on a finite-dimensional normed space is continuous. In infinite-dimensional spaces, linearity alone does not guarantee continuity.\n\nDependence on the norm\n\nConsider endpoint evaluation on \\(C([0,1])\\):\n\n\\[E_0(f)=f(0).\\]\n\nIt is continuous under the uniform norm. It is not continuous under the norm\n\n\\[\\|f\\|_1=\\int_0^1|f(t)|\\,dt.\\]\n\nIndeed, the continuous functions\n\n\\[f_n(t)=\\max(1-nt,0)\\]\n\nsatisfy\n\n\\[\\|f_n\\|_1=\\frac{1}{2n}\\longrightarrow 0, \\qquad E_0(f_n)=1.\\]\n\nThus, continuity is a property of both the functional and the structure placed on its domain, not merely of its formula."
    },
    {
      "heading": "Dual spaces and representation",
      "text": "The collection of all linear functionals on \\(X\\) forms its algebraic dual space. When \\(X\\) has a norm or topology, the continuous dual consists only of the continuous linear functionals. Notation varies: \\(X^*\\) often denotes the continuous dual in analysis and the algebraic dual in purely algebraic contexts.\n\nFor a normed space, the continuous dual is itself a Banach space under the operator norm, even when the original space is not complete.\n\nIn finite-dimensional coordinates, a linear functional has the form\n\n\\[\\ell(x_1,\\ldots,x_n)=a_1x_1+\\cdots+a_nx_n.\\]\n\nFunctionals therefore generalize familiar scalar-valued linear expressions to spaces whose elements may have infinitely many degrees of freedom.\n\nOn a Hilbert space, every continuous linear functional has a unique representation by an inner product. With the convention that the inner product is linear in its first argument, there is a unique \\(y\\in H\\) such that\n\n\\[\\ell(x)=\\langle x,y\\rangle \\qquad\\text{for all }x\\in H,\\]\n\nand \\(\\|\\ell\\|=\\|y\\|\\). This is the Riesz representation theorem for Hilbert spaces.\n\nA related, distinct representation theorem identifies continuous linear functionals on \\(C(K)\\), for compact Hausdorff \\(K\\), with integration against suitable finite regular measures. These results explain why inner products and integrals recur as fundamental examples of functionals."
    },
    {
      "heading": "Differentiation and variations",
      "text": "A nonlinear functional can often be differentiated by studying how its value changes under perturbations of its argument.\n\nFor a real-valued functional \\(J\\) on an open subset of a real normed space, the directional variation at \\(u\\) in a direction \\(h\\) is\n\n\\[\\delta J(u;h) = \\left.\\frac{d}{d\\varepsilon}J(u+\\varepsilon h)\\right|_{\\varepsilon=0},\\]\n\nwhen this derivative exists.\n\nIf these directional derivatives form a bounded linear functional of \\(h\\), they define a Gâteaux derivative under a common convention. Terminology varies slightly between texts. Existence of all directional derivatives alone does not ensure linearity or continuity in the direction.\n\nThe stronger Fréchet derivative requires a bounded linear functional \\(DJ(u)\\) satisfying\n\n\\[J(u+h)=J(u)+DJ(u)[h]+o(\\|h\\|) \\qquad\\text{as }\\|h\\|\\to 0.\\]\n\nFréchet differentiability implies Gâteaux differentiability, but the converse does not hold in general.\n\nFor example, on the real space \\(L^2([a,b])\\),\n\n\\[J(u)=\\int_a^b u(t)^2\\,dt\\]\n\nhas derivative\n\n\\[DJ(u)[h]=2\\int_a^b u(t)h(t)\\,dt.\\]\n\nThe derivative at each \\(u\\) is itself a continuous linear functional.\n\nOn a real Hilbert space, this derivative can be represented by a gradient:\n\n\\[DJ(u)[h]=\\langle \\nabla J(u),h\\rangle.\\]\n\nThe derivative is intrinsically a dual-space element; its identification with a gradient depends on the chosen inner product.\n\nEuler–Lagrange equations\n\nA central variational example is\n\n\\[J[u]=\\int_a^b L(t,u(t),u'(t))\\,dt.\\]\n\nFor sufficiently smooth \\(L\\) and \\(u\\), perturbing \\(u\\) by \\(u+\\varepsilon h\\), with \\(h(a)=h(b)=0\\), gives\n\n\\[\\delta J(u;h) = \\int_a^b \\left( \\frac{\\partial L}{\\partial u}h+ \\frac{\\partial L}{\\partial u'}h' \\right)\\,dt.\\]\n\nUsing integration by parts, stationarity for all admissible variations yields the Euler–Lagrange equation:\n\n\\[\\frac{\\partial L}{\\partial u} -\\frac{d}{dt}\\left(\\frac{\\partial L}{\\partial u'}\\right)=0.\\]\n\nStationarity is generally a necessary condition for a smooth unconstrained extremum, not a sufficient condition for a minimum."
    },
    {
      "heading": "Optimization and convexity",
      "text": "In mathematical optimization, a functional often assigns a cost to an entire candidate function. A typical problem is\n\n\\[\\min_{u\\in\\mathcal A}J[u],\\]\n\nwhere \\(\\mathcal A\\) is a set of admissible functions encoding boundary conditions or other constraints.\n\nA real-valued functional is a convex functional if its domain is convex and\n\n\\[J(\\theta u+(1-\\theta)v) \\leq \\theta J(u)+(1-\\theta)J(v)\\]\n\nfor \\(0\\leq\\theta\\leq 1\\).\n\nFor differentiable convex functionals on open convex domains, a vanishing derivative is sufficient for a global minimum. Strict convexity ensures that there is at most one minimizer, but does not by itself establish existence.\n\nExistence in infinite-dimensional problems usually needs additional structure. A common method combines boundedness of minimizing sequences, an appropriate weak compactness property, and lower semicontinuity. Merely showing that a functional is continuous and bounded below does not ensure that its infimum is attained."
    },
    {
      "heading": "Applications",
      "text": "Differential equations. Many partial differential equations arise as stationarity conditions for energy functionals. Variational formulations also provide a natural setting for a weak solution, whose derivatives need not exist in the classical sense.\n\nInverse problems. In an inverse problem, a functional may combine mismatch with observed data and a penalty:\n\n\\[J(u)=\\|Au-y\\|_Y^2+\\lambda R(u), \\qquad \\lambda>0.\\]\n\nThe first term measures agreement with the data, while the regularization functional \\(R\\) imposes a mathematical preference such as small norm or smoothness.\n\nStatistics and learning. Population risk assigns a scalar expected loss to a prediction function:\n\n\\[\\mathcal R(f)=\\mathbb E[L(Y,f(X))].\\]\n\nThis is a functional of \\(f\\), rather than merely a function of a finite parameter vector.\n\nGeneralized functions. In distribution theory, a distribution is a continuous linear functional on a specified space of smooth test functions. The Dirac distribution, for example, acts by\n\n\\[\\delta_{t_0}(\\varphi)=\\varphi(t_0).\\]\n\nIts continuity is understood with respect to the topology of the test-function space."
    },
    {
      "heading": "Historical development",
      "text": "Functionals emerged from problems in which the unknown was a curve or function rather than a finite list of numbers. Eighteenth-century work on variational problems, especially by Euler and Lagrange, developed techniques for studying quantities such as length, action, and energy.\n\nDuring the late nineteenth and early twentieth centuries, the study of mappings on function spaces became more systematic. Work by Vito Volterra, Jacques Hadamard, and other mathematicians helped develop this viewpoint. The later abstract theories of normed and topological vector spaces placed linear functionals, duality, and continuity within the general framework of functional analysis."
    }
  ],
  "links": [
    "mathematical-function",
    "real-number",
    "complex-number",
    "vector-space",
    "functional-analysis",
    "calculus-of-variations",
    "lp-space",
    "integral",
    "norm-mathematics",
    "expected-value",
    "linear-functional",
    "linear-map",
    "normed-vector-space",
    "dual-space",
    "banach-space",
    "hilbert-space",
    "inner-product",
    "gradient",
    "integration-by-parts",
    "euler-lagrange-equations",
    "mathematical-optimization",
    "convex-function",
    "partial-differential-equation",
    "weak-solution",
    "inverse-problem",
    "regularization"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}