Documentation

TauCeti.RingTheory.Huber.Completion

The completion of a Huber ring #

The Hausdorff completion  of a Huber ring A is again a Huber ring. If (A₀, I) is a pair of definition for A, then the closure Â₀ of the image of A₀ is an open subring of Â, and the ideal Î of Â₀ generated by the image of I is an ideal of definition for it.

That Huber structure rests on one identity: the n-th power Îⁿ is exactly the closure of the image of Iⁿ (TauCeti.Huber.PairOfDefinition.completionIdeal_pow). Those closures are a neighbourhood basis of zero in Â, so the identity gives both halves of IsAdic Î at once. Read from right to left it also makes the extension of an open ideal of A to  open: the closure of the image of Iⁿ is the ideal of Â₀ generated by that image, so an ideal of  containing the image already contains that closure, which is an open neighbourhood of zero.

The inclusion Îⁿ ⊆ closure (Iⁿ) is formal. The reverse inclusion is the substantial point, and it is the only place where finite generation of I enters the proof of completionIdeal_pow: fixing a finite family G generating Iⁿ, an element of the closure is approximated to order n + k, for every k, by a combination of the same family G with coefficients in Iᵏ. Finite generation is used once more, and directly, to see that Î is itself finitely generated (TauCeti.Huber.PairOfDefinition.fg_completionIdeal). The coefficients accumulate to series which converge because  is complete, and their sums exhibit the element as a combination of G.

Main definitions #

Main results #

References #

The closure in  of the image of Iⁿ, as an additive subgroup. These closures are the neighbourhood basis of zero of the completion, and they turn out to be the powers of the ideal of definition of the completion (completionIdeal_pow).

Equations
Instances For

    The closure of the image of Iⁿ is open in Â.

    The image of Iⁿ lies in its closure.

    @[simp]

    An element of A lies in the closure of the image of Iⁿ in  exactly when it lies in the image of Iⁿ. There is no separation hypothesis on A, so this holds even when A →  is not injective; the implication from right to left is coe_mem_completionIdealImage.

    The closures of the images of the Iⁿ decrease with n.

    The closure of the image of Iⁿ lies in the ring of definition of the completion.

    Wedhorn Remark 6.8: the closures of the images of the powers of the ideal of definition are a neighbourhood basis of zero in the completion.

    A set is a neighbourhood of x in  exactly when it contains the translate by x of one of the closures completionIdealImage n. This is hasBasis_nhds_zero_completion at the point x.

    Approximation from A. Every point of  lies within completionIdealImage n of the image of some element of A, for each n: the image of A is dense, and completionIdealImage n is a neighbourhood of zero.

    The closure of the image of Iⁿ absorbs multiplication by the ring of definition of the completion, because Iⁿ is an ideal of A₀ and multiplication is continuous.

    The closure of the image of Iⁿ, viewed as an ideal of the ring of definition of the completion.

    Equations
    Instances For
      @[simp]

      Membership in the ideal cut out by the closure of the image of Iⁿ.

      The completion map, restricted to a ring of definition and corestricted to the ring of definition of the completion.

      Equations
      Instances For
        @[simp]

        The corestricted completion map agrees with the completion map on A₀.

        The ideal of definition of the completion: the ideal of Â₀ generated by the image of I.

        Equations
        Instances For
          @[simp]

          Membership in Î is membership in the ideal spanned by the image of I.

          The ideal of definition of the completion is finitely generated.

          Wedhorn Remark 6.8: the n-th power of the ideal of definition of the completion is exactly the closure of the image of Iⁿ.

          Wedhorn Remark 6.8: the subspace topology on the ring of definition of the completion is the adic topology of the ideal of definition of the completion.

          Wedhorn Remark 6.8: a pair of definition of A induces one of the completion of A, with the closure of the image of A₀ as ring of definition and the ideal generated by the image of I as ideal of definition.

          Equations
          • One or more equations did not get rendered due to their size.
          Instances For
            @[simp]

            Membership in the ideal of definition of the completed pair is membership in I · Â₀.

            Stated as a membership characterisation rather than an equation because the type of idealOfDefinition depends on ringOfDefinition, which the opaque body of completion does not expose.

            An ideal of  that contains the image of Iⁿ contains the whole closure of that image. The hypothesis constrains J only at the points of  that come from A.

            The extension of an open ideal along the completion map is open: for an open ideal J of a Huber ring A, the ideal of  generated by the image of J is open.

            The image of a bounded subset of A is bounded in Â.

            Stated for a Huber ring rather than for a chosen pair of definition: the conclusion does not mention one, so the choice is internal to the proof.

            The closure argument follows AINTLIB (github.com/CBirkbeck/AINTLIB, Apache-2.0, branch dev/adic-spaces at 37bbdaeb9), projects/AdicSpaces/Adic spaces/LocalizationTopology.lean:545, where it is stated for the rational-localisation completion. No proof was copied.

            A power-bounded element of A stays power-bounded in Â.

            The converse of isPowerBounded_completion_coe_of_isPowerBounded. An element of A whose image in  is power-bounded was already power-bounded.

            @[simp]

            Wedhorn Lemma 7.47(1), at the level of elements: A⁰ is the preimage of Â⁰. An element of A is power-bounded exactly when its image in the completion is.

            Wedhorn Remark 6.8: the completion of a Huber ring is a Huber ring.

            Wedhorn Remark 6.8: the completion of a Tate ring is a Tate ring.

            Completion preserves rings of integral elements (Wedhorn, Lemma 7.47): the closure in  of the image of a ring of integral elements of a Huber ring A is a ring of integral elements of Â.