Documentation

Atlas.AlgebraicTopologyI.code.Section13

theorem AlgebraicTopologyI.lebesgue_covering_lemma {M : Type u_1} [PseudoMetricSpace M] [CompactSpace M] {ι : Sort u_2} {U : ιSet M} (hU_open : ∀ (i : ι), IsOpen (U i)) (hU_cover : Set.univ ⋃ (i : ι), U i) :
ε > 0, ∀ (x : M), ∃ (i : ι), Metric.ball x ε U i

Lemma 13.3 (Lebesgue covering lemma). Let M be a compact metric space and let U : ι → Set M be an open cover. Then there is ε > 0 such that for every x ∈ M, the ball B_ε(x) is contained in some U i.

noncomputable def SubdivisionDiameter.barycenter {E : Type u_1} [SeminormedAddCommGroup E] [NormedSpace E] (n : ) (v : Fin (n + 1)E) :
E

The barycenter of n + 1 points v 0, …, v n in a real normed space: the average (n + 1)⁻¹ · Σ v i.

Instances For
    theorem SubdivisionDiameter.barycenter_sub_vertex {E : Type u_1} [SeminormedAddCommGroup E] [NormedSpace E] (n : ) (v : Fin (n + 1)E) (i : Fin (n + 1)) :
    barycenter n v - v i = (↑(n + 1))⁻¹ j : Fin (n + 1), (v j - v i)

    The displacement from a vertex v i to the barycenter expressed as the scaled sum of all vertex displacements v j - v i.

    The barycenter of vertices v 0, …, v n lies in the convex hull of range v.

    theorem SubdivisionDiameter.dist_barycenter_vertex_le {E : Type u_1} [SeminormedAddCommGroup E] [NormedSpace E] (n : ) (v : Fin (n + 1)E) (i : Fin (n + 1)) :
    dist (barycenter n v) (v i) n / (n + 1) * Metric.diam (Set.range v)

    The distance from the barycenter of vertices v 0, …, v n to any vertex v i is bounded by n / (n + 1) times the diameter of the vertex set. This is the key estimate used in the proof of Lemma 13.1.

    theorem SubdivisionDiameter.dist_barycenter_convexHull_le {E : Type u_1} [SeminormedAddCommGroup E] [NormedSpace E] (n : ) (v : Fin (n + 1)E) (p : E) (hp : p (convexHull ) (Set.range v)) :
    dist (barycenter n v) p n / (n + 1) * Metric.diam (Set.range v)

    The distance from the barycenter of vertices v 0, …, v n to any point of the convex hull conv(range v) is bounded by n / (n + 1) times the diameter of the vertex set.

    inductive SubdivisionDiameter.IsSubdivSimplex {E : Type u_1} [SeminormedAddCommGroup E] [NormedSpace E] {n : } :
    (Fin (n + 1)E)Set EProp

    Inductive predicate IsSubdivSimplex v τ characterising the vertex sets τ ⊆ E of simplices arising in the barycentric subdivision of the affine simplex with vertices v. The base case is a single point; the cone step adjoins the barycenter of v to a face subdivision.

    Instances For
      theorem SubdivisionDiameter.range_comp_succAbove_subset {E : Type u_1} {n : } (v : Fin (n + 2)E) (k : Fin (n + 2)) :

      Removing the k-th vertex via Fin.succAbove k produces a face whose vertex set is a subset of the original.

      Every subdivision simplex of an affine simplex σ with vertices v is contained in the convex hull of range v, i.e. in σ itself.

      Every subdivision simplex is a bounded subset of E.

      theorem SubdivisionDiameter.diam_face_le {E : Type u_1} [SeminormedAddCommGroup E] {n : } (v : Fin (n + 2)E) (k : Fin (n + 2)) :

      The diameter of any face of the affine simplex with vertices v is at most the diameter of range v.

      theorem SubdivisionDiameter.diam_subdivSimplex_le {E : Type u_1} [SeminormedAddCommGroup E] [NormedSpace E] {n : } {v : Fin (n + 1)E} {τ : Set E} :
      IsSubdivSimplex v τMetric.diam τ n / (n + 1) * Metric.diam (Set.range v)

      Lemma 13.1 (subdivision diameter bound). For any subdivision simplex τ of the affine n-simplex σ with vertices v, diam(τ) ≤ n / (n + 1) · diam(σ).

      The k-fold composition of a chain-complex endomorphism f : C ⟶ C, i.e. f^k. Used to express iterates of the subdivision operator $^k : S_*(X) → S_*(X) from Lemma 13.4.

      Instances For
        @[simp]

        The 0-th iterate of f is the identity.

        @[simp]

        The (n + 1)-th iterate of f factors as f followed by f^n.

        Lemma 13.4 (chain-level witness). For every k, the k-fold iterate $^k of the subdivision operator is chain-homotopic to the identity on S_*(X). The witness is built by induction on k, composing the one-step subdivision chain homotopy with the previous iterate.

        Instances For

          Lemma 13.4. For every k ≥ 0, $^k ≃ 1 : S_*(X) → S_*(X) as chain maps, i.e. there exists a chain homotopy between the k-th iterate of the subdivision operator and the identity.