Alphabeta Math
LemmaStatement: AI-generatedProof: AI-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Supported perfect complexes are invariant under an open neighbourhood

PDF p. 7: Zariski locality; PDF p. 15: Localization diagram

Statement

Assume the Axiom of Choice and fix the suitable universes of the supplied derived and K-theory constructions. Let X be qcqs, let T⊆X be closed, and let v:V↪X be a quasi-compact open containing T. Actual restriction induces an exact equivalence v∗:PerfT(X)→∼PerfT(V), with inverse given in the supplied sheaf-derived model by degreewise module extension by zero v!. It induces the natural equivalence KT(X)≃KT(V), compatible with the support-to-global inclusions and restriction on global K spectra. No quasi-compactness of X∖T is needed for this categorical excision lemma.

Facts & Assumptions

[A1]

Choice and the universe bounds are used to invoke the enhanced perfect-category and K-theory suppliers. The Axiom of Choice

Given: X,T,V,v as stated and the supplied sheaf-derived enhancement.

Goal: Construct an actual exact inverse on supported perfect categories, not merely a triangulated equivalence.

[F1]

External exact extension by zero for module sheaves gives actual degreewise exact functors v!,v∗, their natural unit/counit, the stalk formula, and v∗v!=id.

[F2]

Perfect complexes on schemes and their support defines perfection locally and support by cohomology stalks; External input on derived enhancements supplies the actual infinity-localizations. Nonconnective K theory of perfect complexes supplies K on these full categories and actual restriction maps.

Proof

1.1F1F2

Let F∈PerfT(V). Exactness and the stalk formula in [F1] show that the cohomology of v!F equals that of F on V and is zero off V, so its support is contained in T. On the open V it is perfect because v∗v!F=F; on the open X∖T it is acyclic, hence equivalent to the zero perfect object. These opens cover X because T⊆V. Thus local perfection in [F2] makes v!F perfect on X. This is why closed ambient support, rather than preservation of arbitrary quasi-coherent sheaves by v!, suffices.

2.1F1F2step 1.1

The degreewise unit identifies v∗v! with the identity. For E∈PerfT(X) the actual counit v!v∗E→E is a quasi-isomorphism: at points of V it is an isomorphism on complex stalks, and outside V the source is zero while the target is acyclic since those points lie outside T. Hence this natural transformation becomes an equivalence in the supplied infinity-localization. Both functors are exact by [F1], and step 1.1 verifies their restrictions have precisely the claimed perfect/support domain. They therefore give mutually inverse exact functors of the actual enhanced full subcategories.

3.1F1F2step 2.1∎

Apply the functor in [F2] to this exact equivalence. The square formed by the two support inclusions and the global/support restriction functors commutes already on complexes, and the common-family realization of the supplied interface represents the same restriction. Thus the K equivalence is compatible with the claimed maps. Extension by zero is used solely as an inverse on these support categories, without requiring it to preserve the common K-flat presentation or to be a global scheme pullback.

Depends on

Used by · 1

Dependency tree · 3 levels

Arrows run from prerequisites to the items that use them. External source nodes link directly to the cited text. The view shows up to three dependency steps; open a prerequisite to continue.