Alphabeta Math
AlgorithmAlgorithm: AI-adaptedProof: 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.

Projective code-manifold syndrome check

PDF p. 15: Paragraph before Eq. (S4.5) and Eqs. (S4.5), (S4.7)–(S4.9); PDF p. 16: First paragraph

Definition

Adopt the finite lifted model and all hypotheses of its first-order fast-check channel, including the Axiom of Choice, Countable Choice and Dependent Choice. Its physical space is the orthogonal sum of the blocks spanned by the basis vectors ∣J,m⟩ with J=0,…,N and m=−J,…,J. A density operator is a positive semidefinite operator of trace one. Let J0∈{1,…,N}, let P=ΠJ0 be the orthogonal projector onto the J0 block, and let the input density operator satisfy Pρ0P=ρ0. Choose a check interval τ≥0. The exact model channel for that interval is Nτ.

Measure the two projectors P and Q=I−P on the output of Nτ. The ideal measurement instrument has unnormalized outputs Bnull=PNτ(ρ0)P and Bdecay=QNτ(ρ0)Q. For either outcome a, its probability is pa=Tr⁡Ba; if pa>0, retain the conditional state Ba/pa. A zero-probability outcome has no conditional state.

For the first-order branch description, put Knull=I+τG and Kδ=τLδ for δ∈{−1,0,1}, using the model's diagonal G and downward jumps Lδ. Replace Nτ by the displayed first-order Kraus sum. The resulting unnormalized null branch is Knullρ0Knull† and the resulting unnormalized decay branch is ∑δKδρ0Kδ†; normalize a branch by its trace only when that trace is positive. Here γ>0 is the inherited equal-rate scale, and for the normalized excited-block input the first-order decay-branch weight is Tr⁡(∑δKδρ0Kδ†)=γτ. Thus for τ>0 its conditional state in this approximation is ∑δKδρ0Kδ†/(γτ); at τ=0 this branch has zero weight and no conditional state. These are branch expressions for the approximation, which need not itself be trace preserving at finite τ.

If an encoded input and a suitable decoder are supplied, apply that decoder after the decay outcome. Here a decoder is a quantum channel from the post-measurement physical space to the logical space; suitable means that it recovers the encoded information for the specified transition-error channel under the stated code assumptions. The check itself does not construct such a decoder or establish that an arbitrary code has one. For an initial state supported in the J0=0 block, the model has no downward jump, and its density operator remains unchanged under the exact evolution.

Facts & Assumptions

[F1]

In the finite lifted model, G is diagonal in the block basis, each Lδ maps an excited block into the next lower block, and every Lδ annihilates the ground block. The exact channel differs from the first-order Kraus sum by O(τ2) in the stated induced norm at fixed model parameters. First-order fast-check channel

[F2]

A finite instrument consists of completely positive outcome maps whose sum is trace preserving; its outcome weight is the trace of the unnormalized output, and its conditional state is the normalized output when that weight is positive. Finite instruments and conditional states

[F3]

An isometric encoding defines a code subspace, and a decoded recovery is a channel from the physical output to the logical space. Code restriction and decoded channel types

[F4]

A recovery cycle specifies its measurement and outcome-dependent decoder; calibration alone does not establish a recovery guarantee. Calibrated recovery cycles and accumulated-error contract

[A1]

Assume the Axiom of Choice, Countable Choice and Dependent Choice, as required by the finite lifted-model supplier. The Axiom of Choice The Axiom of Countable Choice (AC_ω) The axiom of dependent choice: a relation in which every element is related to something admits an ℕ-indexed chain

[F5]

For an orthonormal basis of a subspace of a finite-dimensional inner-product space, the orthogonal projection is the sum of the corresponding basis components. It is linear, has that subspace as image and its orthogonal complement as kernel, is idempotent, and its complement is the projection onto the orthogonal complement. Orthogonal projection is linear, and an orthonormal basis (eᵢ) of W gives P_Wv=∑ᵢ⟨ v,eᵢ⟩ eᵢ

[F6]

A finite projective measurement is a finite family of self-adjoint idempotent endomorphisms that are pairwise orthogonal and sum to the identity. Finite projective measurements

[F7]

For compatible finite rectangular matrices over a field, Tr⁡(AB)=Tr⁡(BA). For A∈ M_m× n(F) and B∈ M_n× m(F), tr(AB)=tr(BA)

[F8]

Under the inherited equal-rate hypotheses, γ>0 and ∑δLδ†Lδ=γΠexc, where Πexc=∑J=1NΠJ is the orthogonal projection onto the excited blocks. Normalized jumps give a scalar total decay operator

Justification

Given: The stated definitions, hypotheses and cited interfaces.

Goal: Justify the stated construction and its scope.

1.1 In the orthonormal block basis, [F5] makes P diagonal with entry one on the J0 block and zero elsewhere. Thus P∗=P, P2=P, and Q=I−P is the complementary orthogonal projector, with Q∗=Q, Q2=Q, PQ=QP=0 and P+Q=I. The pair is a finite projective measurement by [F6]. Each map X↦PXP or X↦QXQ is completely positive because conjugation by the corresponding projector remains positive after every finite identity amplification. By [F7] and idempotence, Tr⁡(PXP)+Tr⁡(QXQ)=Tr⁡(XP)+Tr⁡(XQ)=Tr⁡X for every endomorphism X. Thus these maps form an instrument in the sense of [F2], and the stated probabilities and conditional states follow. The exact channel precedes this instrument, so its two probabilities sum to one. [given, F1, F2, F3, F4, F5, F6, F7]

2.1 By [F1], the diagonal Knull preserves the J0 block and the jumps take it into the orthogonal J0−1 block. Consequently PKnullρ0Knull†P=Knullρ0Knull†, while projecting any jump term with P gives zero. Conversely, projection with Q removes the null term and leaves the jump terms. The approximation bound in [F1] concerns unnormalized channel outputs; it does not by itself give a uniform bound on normalized conditional states when an outcome probability is small. [F1, F2, F3, F4, step 1.1]

3.1 Since J0≥1, the supported input satisfies Πexcρ0=ρ0 and Tr⁡ρ0=1. By the finite trace cyclicity in [F7], Tr⁡(∑δKδρ0Kδ†)=τ∑δTr⁡(ρ0Lδ†Lδ)=γτTr⁡(ρ0Πexc)=γτ using [F8]. This is the weight of the approximate decay branch selected in step 2.1. Since γ>0, it is positive precisely when τ>0, giving the stated division by γτ; at τ=0 every Kδ is zero. The exact branch probability remains defined by the exact instrument, and is not identified with this first-order weight. [given, F1, F2, F7, F8, step 2.1, algebra]

4.1 The decoder is an additional input of the type in [F3], as in the outcome-dependent cycle of [F4]. Its suitability is a separate hypothesis, so the procedure makes no unconditional correction claim. On the one-dimensional ground block the jumps vanish by [F1], and the diagonal Hamiltonian acts as a scalar; its unitary phase cancels in the density operator. Hence the exact ground-block state is unchanged. [F1, F2, F3, F4, step 1.1] ∎

Depends on

Used by · 5

Dependency tree · 2 levels

Each result is drawn at its shortest distance from this item. Arrows run from prerequisites to the items that use them; this item has a heavier outline. External source nodes link directly to the cited text. The view shows up to two dependency steps; open a prerequisite to continue.

Sources