{
  "id": "hilberts-problems",
  "title": "Hilbert’s Problems",
  "lang": "en",
  "source_lang": "en",
  "summary": "Hilbert’s problems are 23 mathematical challenges published in 1900 that shaped research across twentieth-century mathematics and remain partly unresolved.",
  "category": "mathematics",
  "subcategory": "discrete-logic",
  "type": "concept",
  "aliases": [
    "Hilbert’s 23 problems",
    "Hilbert problems"
  ],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/hilberts-problems/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/hilberts-problems/",
      "json": "https://aiwiki.page/zh/hilberts-problems.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Hilbert’s problems are a collection of 23 mathematical problems presented by David Hilbert in 1900 as challenges for the coming century. They range from the foundations of mathematics and the structure of space to prime numbers, algebraic equations, and differential equations. Some ask for a proof of a precise conjecture; others propose an extensive research programme. Their influence came not only from the answers eventually obtained but also from the theories and methods developed in pursuing them.\n\nHilbert introduced the programme in an address to the second International Congress of Mathematicians in Paris on 8 August 1900. The published text contained all 23 problems, although the oral lecture discussed only a selection. Several have been solved affirmatively, others refuted, and some shown to depend on the axioms or interpretations adopted. Important parts remain open, notably questions involving the Riemann hypothesis, the distribution of primes, and the topology of algebraic curves and differential equations.\n\nThere is consequently no universally satisfactory count of how many Hilbert problems have been “solved.” A single numbered problem may contain several questions, and a negative answer, an independence theorem, and the completion of a broad research programme represent different kinds of mathematical achievement."
    },
    {
      "heading": "The Paris address and its intellectual setting",
      "text": "Hilbert delivered his address, usually known in English as “Mathematical Problems,” at a moment when mathematics was becoming both more specialized and more interconnected. Set theory had introduced new conceptions of infinity; algebraic methods were transforming geometry; and questions about rigor had become central to analysis. Hilbert’s own work crossed these boundaries. His research on polynomial invariants, algebraic number fields, and the axiomatic foundations of geometry supplied several of the list’s immediate concerns.\n\nThe address was not simply a catalogue of difficult exercises. Hilbert argued that important problems should be understandable, challenging, and capable of stimulating new methods. He also emphasized that apparently remote branches of mathematics could illuminate one another. A geometric construction might depend on algebra, while a question about whole numbers could require complex analysis. The arrangement of the list reflected this outlook: foundational questions came first, followed by geometry and physics, arithmetic and algebra, and finally analysis.\n\nSeveral entries arose from recent work rather than from centuries-old conjectures. The third problem followed debates about volume and geometric dissection. The seventh concerned transcendence questions suggested by the distinction between algebraic and transcendental numbers. The fourteenth grew from Hilbert’s own finiteness theorems. Other entries incorporated established questions, including the Riemann hypothesis and the extension of Kronecker’s ideas about algebraic number fields.\n\nThe original German publication appeared in 1900, and an English translation by Mary Winston Newson appeared in 1902. These publications made the complete list available beyond the audience of the Paris lecture. Its influence subsequently depended on teaching, research surveys, and repeated reassessments by specialists, rather than on any institution formally administering the problems.\n\nHilbert also recorded a proposed twenty-fourth problem in his private notes, concerning criteria for the simplicity of proofs and a general theory of methods of proof. It was not part of the published list and became widely known only after Rüdiger Thiele’s historical investigation, published in 2003. Its existence helps distinguish Hilbert’s working ideas from the canonical collection: “Hilbert’s problems” conventionally means the 23 published problems, not every question Hilbert considered proposing."
    },
    {
      "heading": "Problems 1–2: infinity, consistency, and the limits of axioms",
      "text": "The first problem concerns the continuum hypothesis, formulated by Georg Cantor. In modern notation, it asserts that no set has a cardinality strictly between that of the natural numbers and that of the real numbers. Equivalently,\n\n\\[2^{\\aleph_0}=\\aleph_1.\\]\n\nHilbert also discussed the question of well-ordering the real numbers. These questions belong to the foundations of set theory, where apparently simple statements about size depend on the underlying axioms.\n\nThe decisive outcome was independence rather than an ordinary proof or disproof. Kurt Gödel showed in 1938 that the continuum hypothesis and the axiom of choice are consistent with Zermelo–Fraenkel set theory if that theory is itself consistent. Paul Cohen showed in 1963 that the negation of the continuum hypothesis is also consistent with the usual axioms, again under an appropriate consistency assumption. Thus Zermelo–Fraenkel set theory with choice, usually abbreviated ZFC, does not decide the continuum hypothesis, provided ZFC is consistent.\n\nThe qualification is essential. Independence does not mean that the statement is meaningless, nor that every possible foundational framework must leave it undecided. Additional axioms can imply or contradict it. Research therefore continues on which extensions of set theory are mathematically fruitful and what kinds of justification they possess. The well-ordering component has a different history: Ernst Zermelo proved in 1904 that the axiom of choice implies that every set can be well-ordered.\n\nThe second problem asks for a proof that the axioms of arithmetic are consistent. Hilbert wanted a direct mathematical demonstration that the formal rules governing numbers could never produce a contradiction. This concern later became central to Hilbert’s programme, which sought to secure substantial parts of mathematics through formalization and suitably elementary reasoning about proofs.\n\nGödel’s incompleteness theorems of 1931 fundamentally changed the question. Under the standard hypotheses, a consistent, effectively axiomatized theory containing enough arithmetic cannot prove its own consistency using its own resources. This does not prohibit every consistency proof. It constrains the relationship between the system being justified and the methods used to justify it.\n\nGerhard Gentzen’s 1936 consistency proof for Peano arithmetic illustrates the distinction. It uses transfinite induction up to the ordinal \\(\\varepsilon_0\\), a principle not provable in Peano arithmetic itself. The result became foundational for proof theory, but it did not provide the uncontroversially finitary justification Hilbert had originally sought. The second problem is therefore best understood through these precise limitations and relative achievements, rather than through an unqualified label such as “solved” or “impossible.”"
    },
    {
      "heading": "Problems 3–5: dissection, straight lines, and continuous groups",
      "text": "The third problem asks whether polyhedra of equal volume can always be cut into finitely many polyhedral pieces and reassembled into one another. Its background was a contrast with plane geometry: polygons of equal area are equidecomposable by finite dissection. Hilbert wanted to know whether volume in three dimensions could be treated in an analogous way, without relying on limiting arguments.\n\nMax Dehn answered negatively in 1900. He constructed what is now called the Dehn invariant, an additive quantity involving edge lengths and dihedral angles that is unchanged by finite dissection and reassembly. Equal volume does not imply equal Dehn invariant. In particular, a cube and a regular tetrahedron of the same volume are not scissors congruent. The problem thus separated numerical volume from the finer equivalence relation of scissors congruence. Later work by Jean-Pierre Sydler showed that volume and the Dehn invariant together classify scissors congruence for ordinary three-dimensional polyhedra.\n\nThe fourth problem asks for geometries in which straight lines provide shortest paths. Its wording permits several interpretations, depending on what regularity is imposed on distance and whether distances must be symmetric. In one formulation, it seeks metrics on suitable regions of projective space whose unparameterized geodesics are straight-line segments. Ordinary Euclidean distance is one example, but it is far from the only one.\n\nA useful additional example is the Hilbert metric on a bounded convex domain. Its distance is defined using the cross-ratio of the two points and the intersections of their line with the domain boundary. Straight segments are shortest paths, but the metric need not come from an inner product. Herbert Busemann and later researchers developed broad approaches to the problem; A. V. Pogorelov obtained important classifications under specified hypotheses. Its status depends on the precise category of metrics intended, rather than on a single universal theorem.\n\nThe fifth problem concerns the role of differentiability in the theory of continuous transformation groups. Its best-known modern form asks whether a topological group that is locally Euclidean must admit the structure of a Lie group compatible with its topology. In other words, can smooth structure be recovered from the group operation and the topology instead of being assumed at the outset?\n\nWork by Andrew Gleason, Deane Montgomery, and Leo Zippin in 1952 established the affirmative result for locally Euclidean topological groups. Hidehiko Yamabe subsequently developed a wider structural theory for locally compact groups. These achievements underpin the usual statement that the fifth problem was solved positively. They should nevertheless be distinguished from broader transformation-group questions, including the Hilbert–Smith conjecture, which concern continuous actions on manifolds and are not exhausted by the locally Euclidean group theorem."
    },
    {
      "heading": "Problem 6: axiomatizing physics and deriving macroscopic laws",
      "text": "The sixth problem asks for the mathematical treatment of the axioms of physics, with particular attention to probability and mechanics. Unlike the third or seventh problem, it is not a proposition with a single anticipated proof. It proposes a programme: identify basic assumptions, establish their compatibility, and explain how different levels of physical description are related.\n\nFor probability, an especially influential development was Andrey Kolmogorov’s 1933 axiomatization. A probability model is represented by a probability space \\((\\Omega,\\mathcal F,P)\\), where \\(\\mathcal F\\) is a sigma-algebra of events and \\(P\\) is a nonnegative, countably additive measure with \\(P(\\Omega)=1\\). This framework placed random variables, expectations, and stochastic processes within measure theory. It supplied a durable mathematical foundation, while leaving separate questions about the physical interpretation of probability and the suitability of particular models.\n\nHilbert also singled out the passage from atomistic descriptions to continuum mechanics. A gas can be described microscopically through the positions and velocities of many particles, kinetically through a distribution function, or macroscopically through density, velocity, and temperature fields. Connecting these descriptions requires more than writing formally similar equations. It requires control of limits, interaction assumptions, initial conditions, and the scales on which the approximation remains valid.\n\nThe derivation of the Boltzmann equation from particle dynamics is a central example. In 1975, Oscar Lanford proved a rigorous result for hard spheres in the Boltzmann–Grad limit, under specified hypotheses and for a short time interval. The theorem explains how a kinetic description can emerge from deterministic mechanics while also displaying the restrictions of a precise mathematical result. Formal derivations that appear valid over long times can be much harder to justify rigorously.\n\nRelated research studies hydrodynamic limits leading from kinetic equations to equations such as the Navier–Stokes equations. Results depend on the scaling regime, boundary conditions, regularity, and the class of solutions. The formation of singularities, turbulent behaviour, and the emergence of irreversible evolution from reversible microscopic laws present distinct mathematical difficulties.\n\nThe sixth problem therefore remains an active family of questions. Axiomatizations of probability and physical theories are major achievements within its scope, but they do not amount to a final axiomatization of all physics. Its continuing mathematical content lies in explicit connections between theories: determining which assumptions suffice, what limiting equations follow, and how accurately those equations describe the underlying model."
    },
    {
      "heading": "Problems 7–10: transcendence, primes, reciprocity, and algorithms",
      "text": "The seventh problem concerns powers of algebraic numbers. Hilbert asked, in particular, whether \\(\\alpha^\\beta\\) must be transcendental when \\(\\alpha\\) is algebraic, \\(\\alpha\\ne0,1\\), and \\(\\beta\\) is algebraic but irrational. Aleksandr Gelfond and Theodor Schneider independently proved the affirmative result in 1934. For complex exponentiation, the theorem applies to every value of the power. A familiar example is\n\n\\[2^{\\sqrt{2}},\\]\n\nwhich is therefore a transcendental number, despite being constructed from algebraic quantities. The theorem was a major advance beyond earlier proofs concerning particular constants such as \\(e\\) and \\(\\pi\\).\n\nThe eighth problem groups several questions about prime numbers. Its most famous component is the Riemann hypothesis, asserting that every nontrivial zero of the Riemann zeta function has real part \\(1/2\\). This statement controls the size of the error term in approximations to the distribution of primes. Hilbert also discussed Goldbach’s conjecture and the occurrence of prime pairs differing by two, now commonly called twin primes.\n\nThese central questions remain unresolved. They must be distinguished from related theorems. The prime number theorem, proved in 1896, gives the leading asymptotic distribution of primes but does not imply the Riemann hypothesis. Harald Helfgott proved the ternary Goldbach conjecture, concerning sums of three primes, in work completed in 2013; the binary conjecture remains open. Results beginning with Yitang Zhang’s 2013 work establish bounded gaps between primes infinitely often, but do not establish infinitely many gaps of size two.\n\nThe ninth problem asks for general reciprocity laws in algebraic number fields. Classical quadratic reciprocity relates the solvability of congruences modulo different primes. The objective was to understand far more general relationships between arithmetic in a field and the behaviour of its extensions. The development of class field theory, particularly Emil Artin’s reciprocity law in 1927, supplied a general framework for abelian extensions of number fields. Broader nonabelian reciprocity belongs to later research programmes, especially the Langlands programme, and should not be conflated with the classical abelian solution.\n\nThe tenth problem asks for a procedure deciding whether a polynomial equation with integer coefficients has an integer solution. Such an equation is called a Diophantine equation. An input might consist of a polynomial \\(P\\), and the requested procedure would always terminate with a correct answer to\n\n\\[P(x_1,\\ldots,x_n)=0,\\qquad x_i\\in\\mathbb Z.\\]\n\nThe question concerns a single method for all such inputs, not the solvability of a particular equation.\n\nYuri Matiyasevich completed the negative solution in 1970, building on work by Martin Davis, Hilary Putnam, and Julia Robinson. Their combined result shows that computably enumerable sets have Diophantine representations. This makes it possible to encode undecidable computational questions into polynomial equations and proves that the requested general algorithm does not exist.\n\nThe negative answer does not prevent effective methods for restricted families. Linear equations, quadratic forms, and many other special classes admit powerful techniques. Nor does it settle the corresponding decision problem over the rational numbers, which remains open. The choice of the domain—integers, rational numbers, or another ring—is part of the mathematical problem, not a minor change in notation."
    },
    {
      "heading": "Problems 11–12: quadratic forms and explicit class fields",
      "text": "The eleventh problem concerns quadratic forms with algebraic numerical coefficients. A quadratic form is a homogeneous degree-two polynomial, such as\n\n\\[q(x_1,\\ldots,x_n)=\\sum_{i,j}a_{ij}x_ix_j.\\]\n\nQuestions include whether it represents a specified number, whether it has a nonzero zero, and when two forms become equivalent under an invertible linear substitution. Over the real numbers, signature gives a basic classification. Over number fields and their rings of integers, arithmetic introduces substantially richer behaviour.\n\nA central achievement was the local–global theory associated with Helmut Hasse. For a quadratic form over a number field, the existence of a nontrivial zero can be tested in the field’s completions: its real and complex completions and the non-Archimedean completions associated with primes. The Hasse–Minkowski theorem states that the form is isotropic over the number field if and only if it is isotropic over every completion.\n\nThis transforms a global question into a coordinated collection of local ones. For example, determining whether a rational quadratic equation has a nontrivial rational solution can involve tests over \\(\\mathbb R\\) and over the \\(p\\)-adic numbers. The theorem is a model of the local–global principle, but its success must not be generalized indiscriminately: analogous principles fail for many equations of higher degree.\n\nThe distinction between rational and integral variables is also important. A solution over a number field does not automatically yield every desired representation over its ring of integers. Integral quadratic forms involve additional questions of equivalence classes, genera, and representation. Thus the eleventh problem generated a large theory whose achievements cannot be compressed into one elementary classification statement.\n\nThe twelfth problem seeks an explicit construction of abelian extensions of number fields by special values of analytic functions. Its model is the Kronecker–Weber theorem: every finite abelian extension of the rational numbers is contained in a cyclotomic field, generated by roots of unity. These generators arise from values of the exponential function.\n\nFor imaginary quadratic fields, the theory of complex multiplication provides a rich analogue. Special values of modular functions, together with suitable elliptic-function values, generate abelian extensions. The interaction is unusually concrete: analytic values encode algebraic extensions whose Galois groups can be described arithmetically.\n\nHilbert sought comparable constructions for arbitrary number fields. General class field theory establishes the existence and structure of abelian extensions, but does not by itself supply the desired explicit generators through analytic functions. Despite major developments involving complex multiplication and conjectures about special values of \\(L\\)-functions, the twelfth problem remains open in its general form. It illustrates the difference between knowing abstractly which objects exist and constructing them through explicit formulas."
    },
    {
      "heading": "Problems 13–15: equations, invariant rings, and enumerative geometry",
      "text": "The thirteenth problem arose from the solution of algebraic equations by functions of fewer variables. General equations of low degree have familiar formulas, and transformations can simplify a general seventh-degree equation to a form with several independent parameters. Hilbert doubted that its solution could be expressed through superpositions of suitable functions of only two variables.\n\nThe outcome depends critically on the permitted functions. In 1957, Andrey Kolmogorov and Vladimir Arnold established results implying that every continuous function of several real variables can be represented through superpositions involving continuous functions of fewer variables. Kolmogorov’s representation theorem gives a particularly striking form:\n\n\\[f(x_1,\\ldots,x_n) = \\sum_{q=0}^{2n} \\Phi_q\\!\\left(\\sum_{p=1}^{n}\\psi_{pq}(x_p)\\right)\\]\n\nfor continuous functions on an appropriate compact domain, with suitable one-variable functions \\(\\Phi_q\\) and \\(\\psi_{pq}\\).\n\nThese results refuted the obstruction Hilbert anticipated in the continuous setting. They do not automatically supply representations in algebraic or analytic function classes. Consequently, “the thirteenth problem was solved negatively” is accurate only when the relevant interpretation is specified. Questions about algebraic functions and the number of variables necessary in their solution remain distinct.\n\nThe fourteenth problem asks whether certain rings of invariants are finitely generated. It grew from invariant theory, where polynomial expressions unchanged by a group action are studied. Finite generation means that every invariant can be obtained as a polynomial in a fixed finite collection of basic invariants. Hilbert’s earlier finite-basis results made a broad affirmative answer plausible.\n\nMasayoshi Nagata produced a counterexample in 1958, showing that finite generation fails in the generality Hilbert proposed. The negative result did not undermine finite generation in important special cases. Invariant rings for reductive algebraic group actions satisfy powerful finiteness theorems, and these form part of the foundations of geometric invariant theory. The contrast between reductive and nonreductive actions explains why an apparently uniform question separates into sharply different theories.\n\nThe fifteenth problem asks for a rigorous foundation for Hermann Schubert’s enumerative calculus. Enumerative geometry counts geometric objects satisfying incidence conditions: for example, how many lines in projective three-space meet four given lines? Over the complex numbers, with the four lines in general position and interpreted projectively, the answer is two.\n\nSchubert’s methods used controlled degenerations, replacing a configuration with a special one in which counting becomes easier. The difficulty was to justify why the resulting count remained valid and how multiplicities or exceptional configurations should be treated. Modern intersection theory supplies rigorous tools for these questions. Schubert calculus studies distinguished geometric classes on Grassmannians and related spaces, while intersection products encode incidence conditions.\n\nThe problem’s resolution is therefore distributed across the development of algebraic geometry, rather than confined to a single proof. Rigorous counting requires specified parameter spaces, suitable genericity assumptions, and correct multiplicities. Once these are fixed, calculations that once relied on geometric intuition can be expressed as precise identities in intersection theory."
    },
    {
      "heading": "Problem 16: real algebraic curves and limit cycles",
      "text": "The sixteenth problem has two major parts. The first concerns the possible shapes and relative positions of the real components of algebraic curves and surfaces. A real plane algebraic curve is defined by a polynomial equation \\(F(x,y)=0\\), or by its homogeneous counterpart in the real projective plane. Although complex geometry strongly constrains its algebraic structure, the topology of its real points can vary substantially.\n\nFor a nonsingular projective plane curve of degree \\(d\\), the complex genus is\n\n\\[g=\\frac{(d-1)(d-2)}{2}.\\]\n\nHarnack’s inequality bounds the number of connected components of the real locus by \\(g+1\\). That bound does not determine which arrangements are possible. Components may be nested or unnested, and their embedding in the projective plane imposes further restrictions. Hilbert was particularly interested in such configurations, including those of degree-six curves.\n\nTwentieth-century work revealed that topology, complex geometry, and arithmetic restrictions interact strongly. Gudkov’s work on sextic curves and Rokhlin’s complex-orientation formulas are prominent examples. Viro’s patchworking method provides a systematic way to construct real algebraic varieties with prescribed features. Nevertheless, there is no complete classification of all possible arrangements in arbitrary degree, and the higher-dimensional questions about algebraic surfaces are broader still.\n\nThe second part concerns limit cycles of planar polynomial differential equations,\n\n\\[\\dot x=P(x,y),\\qquad \\dot y=Q(x,y).\\]\n\nA limit cycle is an isolated periodic orbit: a closed trajectory with no other periodic trajectories arbitrarily close to it. Such cycles represent persistent oscillatory behaviour in models, but Hilbert’s question is mathematical: how many can occur, and how can they be arranged, when the degrees of \\(P\\) and \\(Q\\) are bounded?\n\nTwo finiteness statements must be separated. One asks whether a particular polynomial vector field has only finitely many limit cycles. Another asks for a uniform upper bound depending only on the degree. Work by Jean Écalle and Yulij Ilyashenko established finiteness for each individual planar polynomial vector field. The uniform-boundedness problem remains open in general.\n\nEven quadratic systems exhibit substantial complexity, and no exact general maximum is known. Special families admit stronger results, but these do not settle the unrestricted question. Both parts of the sixteenth problem therefore remain major sources of research: one studies the topology of real zero sets, the other the recurrent dynamics of real differential equations."
    },
    {
      "heading": "Problems 17–18: positive polynomials, symmetry, and packing",
      "text": "The seventeenth problem asks whether every real polynomial that is nonnegative everywhere can be written as a sum of squares of rational functions. It is important that the summands are allowed to be rational functions rather than polynomials. A sum of polynomial squares is necessarily nonnegative, but not every nonnegative polynomial has such a representation.\n\nEmil Artin proved the affirmative answer in 1927. If\n\n\\[f(x_1,\\ldots,x_n)\\geq0\\]\n\nfor every real input, then there exist rational functions \\(r_1,\\ldots,r_m\\) with real coefficients such that\n\n\\[f=r_1^2+\\cdots+r_m^2.\\]\n\nThe equality is an identity in the field of rational functions. Individual summands may have denominators that vanish at some real points, so the theorem is not a claim that each displayed rational expression is defined everywhere.\n\nArtin’s proof linked positivity to the algebraic theory of ordered fields and helped establish real algebraic geometry. The distinction between polynomial and rational sums of squares remains central. For example, the Motzkin polynomial\n\n\\[x^4y^2+x^2y^4+1-3x^2y^2\\]\n\nis nonnegative but is not a sum of squares of real polynomials. Questions about such certificates of positivity now connect with mathematical optimization, although efficient computational certificates require additional degree bounds and structural assumptions.\n\nThe eighteenth problem contains several questions about space. One concerns the finiteness of essentially different crystallographic groups in each fixed dimension. Ludwig Bieberbach established the relevant finiteness theorem in 1911–1912. A crystallographic group acts discretely on Euclidean space with a compact quotient, formalizing the symmetry structure of a periodic crystal.\n\nAnother part asks whether space can be tiled by congruent polyhedra without those tiles forming a single orbit under a symmetry group of the tiling. Karl Reinhardt constructed a three-dimensional example in 1928. This must be distinguished from the later question of aperiodic tilings: failure of tile-transitivity and absence of all translational periodicity are different properties.\n\nThe most familiar part concerns the densest packing of equal spheres. In three dimensions, the Kepler conjecture predicts the density\n\n\\[\\frac{\\pi}{\\sqrt{18}}\\approx0.74048,\\]\n\nachieved by face-centred cubic packing and other close-packed arrangements. Thomas Hales announced a proof in 1998; a full account appeared in 2005. The Flyspeck project completed a formal verification in 2014, providing a machine-checked foundation for the extensive geometric and computational argument.\n\nSphere packing in arbitrary dimension remains much less understood. Exact optimal densities are known only in selected dimensions; the eight- and twenty-four-dimensional solutions announced in 2016 are especially notable. Thus the classical three-dimensional packing question has a precise answer, while the broader high-dimensional programme continues."
    },
    {
      "heading": "Problems 19–23: regularity, boundary values, and analytic representation",
      "text": "The nineteenth problem asks whether solutions of regular variational problems must be analytic. In the calculus of variations, a typical problem seeks a function minimizing an integral such as\n\n\\[\\mathcal E[u]=\\int_\\Omega F(x,u,\\nabla u)\\,dx.\\]\n\nA minimizer initially defined in a weak function space need not visibly possess ordinary derivatives, let alone a convergent power-series expansion. The issue is whether the structure of the minimizing problem forces this additional regularity.\n\nEnnio De Giorgi and John Nash independently obtained decisive regularity results in 1957–1958. They proved Hölder continuity for solutions of broad classes of uniformly elliptic equations with rough coefficients. In the scalar variational setting relevant to Hilbert’s question, this opened the route from weak solutions to higher regularity; with suitable analytic and ellipticity hypotheses, analyticity follows.\n\nThe qualifications are mathematically essential. Regularity depends on ellipticity, growth conditions, whether the unknown is scalar or vector-valued, and the dimensions involved. Counterexamples for certain systems show that the scalar theory cannot simply be transferred to every variational problem. The standard affirmative assessment of the nineteenth problem refers to the regular elliptic framework, not to all minimizers of all functionals.\n\nThe twentieth problem concerns the existence of solutions to boundary-value problems. A boundary-value problem specifies an equation inside a domain together with conditions on its boundary, such as fixed values of the unknown function. Hilbert asked how generally such problems could be solved when the data satisfied appropriate continuity and regularity assumptions.\n\nModern methods include direct minimization, compactness arguments, weak formulations, and the theory of Sobolev spaces. For the Dirichlet problem for Laplace’s equation, for example, energy minimization can produce a weak solution, while additional hypotheses determine its boundary behaviour and smoothness. But existence is not universal: incompatible boundary data, unsuitable domains, nonlinearities, or loss of coercivity can obstruct it. The twentieth problem consequently developed into a collection of precise existence theories rather than one theorem covering every differential equation.\n\nThe twenty-first problem asks for linear differential equations with prescribed monodromy. Analytic continuation of a basis of solutions around singular points transforms that basis by matrices. These transformations form a representation of the fundamental group of the punctured domain. The Riemann–Hilbert problem reverses the process: starting with such a representation, construct a differential equation realizing it.\n\nHere the formulation determines the answer. A broad realization by regular-singular connections is available, but the stronger demand for a Fuchsian system on the Riemann sphere with exactly the prescribed singular points is not always possible. Andrey Bolibrukh produced a counterexample in 1989 to the unrestricted affirmative claim in this stronger form. The result exposed the role of the underlying holomorphic vector bundle and corrected an earlier belief that the classical problem had been completely solved.\n\nThe twenty-second problem concerns the uniformization of analytic relations by functions of one complex variable. Its central achievement is the uniformization theorem, proved through work of Paul Koebe and Henri Poincaré in 1907. Every simply connected Riemann surface is conformally equivalent to the sphere, the complex plane, or the unit disk. General Riemann surfaces can consequently be studied through these universal covering surfaces and appropriate groups of conformal transformations. The theorem connects complex analysis with topology and non-Euclidean geometry.\n\nThe twenty-third problem calls for further development of the calculus of variations itself. It includes questions about existence, differential equations arising from variational principles, and the extension of classical methods. Twentieth-century developments introduced generalized solutions, lower semicontinuity, compactness, and relaxed formulations when classical minimizers fail to exist. In such formulations, the admissible class is enlarged or the functional is replaced by a suitable lower-semicontinuous extension. This can recover an attainable minimum while preserving a precise relationship to minimizing sequences in the original problem."
    }
  ],
  "links": [
    "david-hilbert",
    "mathematics",
    "set-theory",
    "continuum-hypothesis",
    "cardinality",
    "kurt-godel",
    "axiom-of-choice",
    "zermelo-fraenkel-set-theory",
    "hilberts-program",
    "godels-incompleteness-theorems",
    "proof-theory",
    "geodesic",
    "lie-group",
    "physics",
    "probability",
    "probability-space",
    "measure-theory",
    "navier-stokes-equations",
    "transcendental-number",
    "prime-number",
    "riemann-hypothesis",
    "riemann-zeta-function",
    "prime-number-theorem",
    "diophantine-equation",
    "algorithm",
    "algebraic-geometry",
    "mathematical-optimization",
    "calculus-of-variations",
    "boundary-value-problem",
    "sobolev-space",
    "laplace-equation",
    "topology"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}