Other meanings of Uniform space
Mathematics
A uniform space is a mathematical space with a uniform structure generalizing metric spaces. Instead of assigning a numerical distance to each pair of points, it specifies which pairs are uniformly close, allowing rigorous definitions of uniform continuity, Cauchy filters, completeness, and uniform convergence without requiring a particular metric.
A uniform space is a set equipped with a collection of entourages, subsets of the Cartesian square X × X that encode a notion of pairs being close. The collection contains the diagonal, is closed under supersets and finite intersections, is stable under taking inverses, and has the composition property: for every entourage U, some entourage V satisfies V∘V ⊆ U. These requirements express reflexive, symmetric, and consistently composable smallness without selecting a numerical scale.
Each entourage U can be viewed as a family of neighborhoods U[x] = {y : (x,y) ∈ U}. The resulting neighborhoods generate a topology, but distinct uniform structures may induce the same topology. Thus uniform spaces retain information about large-scale consistency of closeness that ordinary topological spaces do not necessarily contain.1
Uniform continuity is the natural morphism between uniform spaces: a map sends every target entourage back to a source entourage, with one source notion of closeness working simultaneously at every point. This is stronger than ordinary continuity, which may use neighborhoods that vary from point to point. Uniformly continuous maps preserve Cauchy filters and Cauchy nets, and they extend uniquely from a dense subspace into a complete uniform space when the relevant hypotheses hold.
Every metric d produces entourages of the form {(x,y): d(x,y) < ε}; metrics that generate the same uniformity have exactly the same uniformly continuous maps and Cauchy behavior. Uniformities can also arise from families of pseudometrics, from topological groups, and from embeddings into products of simpler spaces.2
A uniform structure sits between a bare set and its induced topology: it determines convergence of filters and nets, but also compares points globally through a common entourage.1 A topological space is uniformizable when its topology is induced by some uniformity; completely regular spaces are precisely the topological spaces with this property. The uniformity is unique for compact Hausdorff spaces, while noncompact spaces can support several compatible uniform structures.
This framework clarifies why completeness is not purely topological. Two compatible uniformities may have the same open sets but different Cauchy filters and different completions. In functional analysis, uniform convergence on a domain is itself a uniform-space construction, and in topological groups the group operation and inversion naturally interact with the uniformity.3
Uniform spaces have distinct left and right uniformities on a noncommutative topological group. They agree for abelian groups but can produce different Cauchy notions and completions in the nonabelian case.2 This distinction is easy to miss when examples are drawn mainly from metric or abelian settings.
Uniform spaces also support completion without first choosing a metric. One construction uses minimal Cauchy filters, identifying filters that cannot be separated by the uniformity; another embeds the space into a complete space characterized by a universal extension property. The Samuel compactification and the theory of uniformizable spaces connect uniform structures with compactifications and topological dynamics. Historically, André Weil’s 1937 treatment helped separate uniform ideas from metric formulas, making the subject useful in general topology, topological algebra, and analysis.4
Notation and terminology vary slightly between authors: some define a uniformity through entourages, while others use pseudometrics, covers, or filters as equivalent descriptions.
Help improve the encyclopedia. Reports go straight to the site manager.